TABLE OF CONTENTS

SESSION MA-SI: Computational Radar Imaging

TIME-FREQUENCY BASED RADAR IMAGE FORMATION ............................................... I
Victor Clien, US Nnvul Reseanrck kboraro7, USA
ISAR - RADAR IMAGING OFTARGETS WITH COMPLICATED MOTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
T q v e Sporc NDRE, Noway
A SURVEY ON ISAR AUTOFOCUSING TECHNIQUES ............................... 9
Fubrizio Berizi. Marco Martorellu, De/'/. OJ Inforrnution Engineering, Iroly; Brett Hrgwuod, Dcfrrm Science & Tecltnobgv
Orgmizurion (DSTO),Ausrrulia: Giro Ilulle Mere und Silvi'r Bruscdi, Dept. of Inqimtution Ensincering,Ita/>,
SAR IMAGE FORMATION VIA INVERSION OF RADON TRANSFORMS.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I3
Nicholm Redding. DSTO, Aurrmliu
LONG INTEGRATION FINE RESOLUTION SAR IMAGE FORMATION.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Ngee L n g 7$i and Sze Li Lee, DSO Nutionol Lcrborutories, Singupore
PERFORMANCE EVALUATION OF BACK-PROJECTION AND RANGE MIGRATION ALGORITHMS IN
FOLIAGEPENETRATION RADAR IMAGING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Yibu No, Ho,rgho Sun, k,
Hui L m , Ling Chior 'hi and Hian Lirrr Clion. .'drool of Elecnirol & El<vtmnic Enginee~ing,
Nnnqang Tecl~nologic~~l
Universiy. Singapore
DISTORTION IN THE INVERSE SYNTHETIC APERTURE RADAR IMAGES FROM MOVING TARGETS 25
S K. Wong, G. I l l @
nnd 6 Risebororqk, Drfencc R&D, G n u d u

SESSION MA-L1: Watermarking I

QUANTITATIVE STEGANALYSIS OF BINARY IMAGES 29
Ming Jiang, Nasir Meiiion. Edrhr.ord Wong, Polyteclmic Unive
AUTHENTICATION OF MPEG-4 BASED SURVEILLANCE VIDEO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33
Michael Prrrr,mte/mkis.Tobirrs Oelbuurn nnd Klaus Diepold, Technisclie Universitmt Muenclren, Germmy
ESTIMATING AND UNDOING ROTATION FOR PRINT-SCAN RESILIENT DATA HIDING 39
Kauslul Solanki, Upamanyu Madhow. B. S.Manjumrh and Shiv Chandrasekaran, University of
RADONIRIDGELET SIGNATURE FOR IMAGE AUTHENTICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Zlim Yao and Nasir Rajpoor, Dept. of Compurer Science, Univers
WATERMARKING OF 3D MODELS FOR DATA HIDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Petros Daras, Aristotle University of Thessaloniki, Greece; Dimitrios Zarpa1a.s. Dimirrios Tzovaras, Informatics and Telemarics
Institute, Greece; and Michael G. Sfrinrzir, Arirrotle University of Thessaloniki, Informatics and Telematics Insrirure, Greece
DATA HIDING IN CURVES FOR COLLUSION-RESISTANT DIGITAL FINGERPRINTING ................................ ,51
Hongmei Gou and Min Wu, University ofMaryland, College Park, USA
ESTIMATION OF AITACKERS SCALE AND NOISE VARIANCE FOR QIM-DC WATERMARK EMBEDDING,. .........
R. (Inaldj L Lagendijk and /vu D. Shterev, Delfr Universiry of Technology. The Netherlands
DATA HIDING USING TRELLIS CODED QUANTIZATION ........ ............................ 59
Ersin Esen, TUBITAK Billen, Turkey; and Aydin Alatan, Department of EE Engineering, METU, Turkey

SESSION MA-LZ: Face Recognition

REGULARIZATION STUDIES ON LDA FOR FACE RECOGNITION. ................................................. .63
Juwei Lu, Konstanrinos N. Plataniotis and Anasrasios N. Venetsanopoulos, Bell Canado Multimedia Laboratov, The Edward S.Rogers
Sr Department of Electrical and Computer Engineering, University of Toronto, Canada
A NOVEL PUPIL LOCALIZATION METHOD BASED ON GABOREYE MODEL AND RADIAL SYMMETRY OPERATOR.. .. .67
Peng Yung, Institute of Compuring Teclrnology of ChineseAcademy of Sciences, China; Bo Du, Shiguang Shun. ICT-ISVISION Joinr R&D
Lnboraforj fo Face Recognition, China: and Wen Gao, lnsritute of Computing Technology of Chinese Academy ofSciences, China
FEATURE SELECTION FOR SUBJECT IDENTIFICATION IN SURVEILLANCE PHOTOS .... ........ . . 7 1
Jie Wimg, Konstantinos Plataniotis and Anastasios Venersanopoulos, Universiry of Toronto. Canada
ESTIMATING FACIAL POSE FROM SPARSE REPRESENTATION ................................................... .75
Hankyu Moon and Man Mille< NEC Labs America, USA
FACE RECOGNITION USING RECURSIVE FISHER LINEAR DISCRIMINANT WITH GABOR WAVELET CODING 79
Cheng Xiang, Xiaoan Fan and TongHeng Lee, NUS. Singapore
A SUPERVISED NONLINEAR LOCAL EMBEDDING FOR FACE RECOGNITI ............................. 83
Jian Cheng, Qingshan Liu. Hanqing LA, Institute of Automation, ChineseAcademy inn: and Yen-Wei Clien, University
of the Ryukyus, Jopan
FUSION OF 2D AND 3D DATA IN THREE-DIMENSIONAL FACE RECOGNITION 87
Alexander Bmnstein, Michael Bronstein. Eyal Gordon and Ron Kimmel, Technion - Israel Institute of Technology,Israel

...
111
POSE AND ILLUMINATION COMPENSATION FOR 3D FACE RECOGNITION.. . . . . .91
Sotiris Malmrioris and Micliael Strintzis, Informatics and 7eleinatics Institute, Greece

SESSION MA-W: Video Compression Standards I

ADVANCED BLOCK SIZE SELECTION ALGORITHM FOR INTER FRAME CODING IN H.2MIMPEC-4 AVC. . . . . . . . . . . . . . . 9 5
Andy C. Kt and Gralmm R. Martin. Universip of Warwick, U K
ROBUSTVIDEOTRANSMISSION USING H.264 AND REAL-VALUED RCH FRAMES . . . 99
Azza Ouled Ziid, Michcl Kieffec Chang-Ming h e and Pierre Dulawzel, W-S laboratory, F m
VIDEO ENCODER COMPLEXITY REDUCTION BY ESTIMATING SKIP MODE DISTORTION.. ......................... 103
lain Richardron and Y+, Zlmo, Tlze Robert Gordon University, U K
A NEW COLOR TRANSFORM FOR RGB C O D I N G . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
H w n Mun Kim, W o ~ i - S I dKim and Doe-Sung Clio, Sa~nsa~igAdvanced lnstitrrte of T e c h
RATE-DISTORTION-COMPLEXITYOPTIMIZATION OF FAST MOTION ESTIMATION IN H.264/MPEG-4 AVC.. . . . . . . . . . . . 1 I I
Jesper SI#nriq-Andersen. Milestone Systems A 6 Denmark; S#ren F o ~ l ~ l z ~ ~ nCOM, n ~ e TrcA.
c 1Jniv. Den,na,t Uensiark: m d
Slmnkur Mnnirel AgliirO. COM, Technical Universip of Denmark, Denmark
FROM 8-TAP DCT TO 4-TAP INTEGER TRANSFORM FOR MPEG TO H.2MIAVC TRANSCODING . . . . . . . . . . . . . . . . . . . . 115
Bo Slim, Heiderr-Packard coin poi^^', USA
FAST MODE DECISION FOR INTER PREDICTION IN H.264 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Qiungliai Dai, Dongdorig Zlrir and Rang Ding. ieee inetitbet: China
A LAGRANGIAN OPTIMIZED RATE CONTROL ALGORITHM FOR THE H.264/AVC ENCODER. ........................ 123
M. Mahdi Gltandi and Mulm,mnrd Clranbari, University of Essex, UK

SESSION MA-L4: Biomedical Image Processing: Segmentation and Quantitative Analysis

A N EVOLUTIONARY SNAKE ALGORITHM FOR THE SEGMENTATION OF NUCLEI IN HISTOPATHOLOGICAL IMAGES . . . I27
Mohatnmed Ali Roula. Alinted Bouridane and Fatili Kurugollu, Queen j university of Belfnsr, UK
MULTIFRAME NONRIGID MOTION ANALYSIS WITH ANISOTROPIC SPATIAL CONSTRAINTS:
APPLICATIONS TO CARDIAC IMAGE ANALYSIS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 I
Ken Wong. Huafeng Liu. Hong Kong Universiry of Science and Teclmology, Hang Kong; Albert Sinusns, Yale Univer.ity, USA: and
Pengcheng Shi, Hong Kong Universiry of Science and Teclinolog>i Hong Kong
SEGMENTATION OF MICROSCOPE CELL IMAGES VIA ADAPTIVE EIGENFILTERS. ................................. 135
Saravana Kuiilac Sim Heng Ong, Surendra Rnnganath, Fook Ern Chew and Tan Cliing Ong, National University of Singapore, Singapom
DETECTION AND CLASSIFICATION OF BRIGHT LESIONS IN COLOR FUNDUS IMAGES. . . . . . . . . . . . . . . . . . . I39
Xiaohui Zhang and Opas Clrutatape, Scliool of EEE, W U , Singapore
EVALUATION OF SHADOW CLASSIFICATION TECHNIQUES FOR OBJECT DETECTION AND TRACKING.. . . . . . . . . . . . . 143
John-Paul Renno, James Orwell and Graeme A Jones, Kingston Univerriry, UK
ADAPTIVE GATING IN GAUSSIAN BAYESIAN MULTI-TARGET TRACKING . . . . . . . . 147
Auguste Genovesio, Ziad Eelhassine and Jean-Christophe Olivo-Marin, Institut Pasteur - Qua
AUTOMATIC ASSESSMENT OF MAMMOGRAPHIC POSITIONING ON THE MEDIOLATERAL OBLIQUE VIEW . . . . . . . . . . 151
Sze Man Kwok, Ramaclmndran Chandrasekhar and Yionni Attikiouzel, Centrefor Intelligent Information Processing Systems, Australia
LOCATING THE OPTIC DISK IN RETINAL IMAGES VIA PLAUSIBLE DETECTION AND CONSTRAINT SATISFACTION.. . 155
Emanuele Trucco and Pawan Komot, Heriot Watt University, U K

SESSION MA-LS: Error Resilience / Concealment I

A NEW FAMILY OF EMBEDDED MULTIPLE DESCRIPTION SCALAR QUANTIZERS. ................................. 159
Augustin Ion Gavrilescu, Adrian Munteanu, Jean Comelis and.Peter Schelkens, Dept. of Electronics and Information Processing
(ETROJ. Belgium
ERROR RESILIENCE VIDEO CODING IN H.264 ENCODER WITH POTENTIAL DISTORTION TRACKING. . . . . . . . . . . . . . . . 163
Yuan Wang, Wen Cao, Chinese Academy of Sciences, China: Huifang Sun, Mitsubirhi Electric Research Laboratories, USA; Qingming
H u n g , ChineseAcademy of Sciences, China: and Yan Lu, Microsoft Research Asia, Clrina
ERROR-PROPAGATION REDUCTION IN A BALANCED MULTIPLE DESCRIPTION VIDEO CODER. .................... 167
Marco Fumagalli, Cefriel - Politecnico di Milano, Italy; Nicola Franchi and Rosa Lnncini, CEFRIEL, ltoly
VISIBILITY OF INDIVIDUAL PACKET LOSSES IN MPEG-2 VIDEO 171
Amy Reibmn, AT&T Labs .Research. USA; Sandeep Kanumuri, UC San Diego, USA: V i m y Vnishampayan,AT&T Lnbs - Research,
USA: and Pamela Cosman, UC San Diego. USA
MOTION COMPENSATED SHAPE ERROR CONCEALMENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Guido Schusres Universiry of Applied Sciences of Eastem Switzerland in Rapperswil, Swirzerland: and Aggelos Katsaggelos,
Northwestem University, USA
A FLUID MODEL FOR ERROR PROPAGATION CHARACTERIZATION IN VIDEO CODING ............................ 179
Xiaorning Sun and C:C. Ja.y Kuo, University af Southern Califomia. USA

iv
A GRADIENT BASED APPROACH FOR STEREOSCOPIC ERROR CONCEALMENT . I81
Maftliirrs Kunreer; Sebnsfinn Kmirer; Cnrsren Clemens and Tlionzas Sikom, Technical Univer.~i
RECURSIVE DECODER DISTORTION ESTIMATION BASED ON AR(I) SOURCE MODELING FOR VIDEO.. . . . . . . . . . . . . . 187
Silo Ekmekci md Thonzas Sikorrr, Tedmical IJniversity Berlin, Cennnny

SESSION MA-PI: Image Segmentation: By Color, Texture, and Edge

FAST FACE SEGMENTATION IN COMPONENT COLOR SPACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Jim Jus 'e De Dim. Universiilad de Crrsrilla-LnMancha, $xiin; rind " r i m Gurcia, Universidad Polit hcnicn de Madrid Spain
COLOR SEGMENTATION OF INKED CHARACTERS :APPLICATION TO MEAT TRACEABILITY CONTROL.. . . . . . . . . . . . 195
Michble Gnuiffrs. C m a g r e j France; Clvisrine Femanrle~-Muloign~~. IRCOM SIC,Puitiers, Frunce: Aluin Trimeau, LIGIC:
St etiennr, Frunce: and Chrisfoplte Collewet, Cemagrej France
LEARNING SKIN DISTRIBUTION USING A SPARSE MAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Rojkiran Gotrunrukkol. Old Dominion Univrrsin: USA: and Vijqon Asari. Old D<JIlliniIJI? Univcrisn: USA
COLOR SPACE SELECTION FOR UNSUPERVISED COLOR IMAGE SEGMENTATION BY HISTOGRAM
MULTITHRESHOLDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .203
Ir,urenr Husin, Nico1n.s Vnndeenhm~ke,Lurlovic Macoire rind Jack-Gerard Posmirr, Lalm-moire LA CIS, F,-unce
ROBUST SKIN SEGMENTATION USING NEIGHBORHOOD INFORMATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,207
Javier Rt~iz~del-Solar
ond Rorlrigo Vcrschae, Universidud de Clrile, Clrilc
PROPOSAL OFTHE HYBRID SPECTRAL GRADIENT METHOD TO EXTRACT CHARACTER /TEXT
REGIONS FROM GENERAL SCENE IMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 I
Yoicliiro Buha and Akira Hirosr, Tire Universit? c?f 7bk?:\, J q i o n
IMAGE SEGMENTATION BASED ON HIERARCHICAL MAPPING . . . . . . . . . . . . . 215
Apinzun Junrla nnd Oraclmr Cliitsobkuk. Departinenf of Contixuer Engi , I:T Institutr of
Technology Lndkrabong., Tlzailand
FEATURES EXTRACTION ON COMPLEX IMAGES.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,219
Pierrick Bourgear, Fobrice Meri~~urleau.Parrick Corrin, Lr2i Lnbonrtop, Fmnce: Kennetlr 7bbin. Onk Ridge Nntionnl L n b ~ r m r y ,
USA; and Frederick Truclicter, Le2i Laborafory, France
A SEMI-SUPERVISED SUPPORT VECKIR MACHINE FOR TEXTURE SEGMENTATION . . . . . >. ....................... ,223
Saeid Sanei. King iCollege London. UK; and Tracey Lee, National University of Singupore, Singupore
SCALE AND ROTATION INVARIANTTEXTUREFEATURES FROM THE D U A L - m E COMPLEX WAVELET TRANSFORM . . 227
Edward Lo,Mark Pickring, Michael Frater and John Amold, Universiry College, The University of New Sourh Wales. Ausrrnlia
SEGMENTATION OF OBJECT REGION USING DEWH INFORMATION. . . . . . . . . . . . . . . . . 23 1
Nak H. Kim and Jai Song Park, Hankuk Univ. of Foreign Studies, Korea
A 3-D CA-BASED EDGE OPERATOR FOR 3-D IMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .235
Sarrra Wongrhanavnsu, Khon Kaen University, Thailand; and Chidchanok Lursinsap, Chulalongkom University, Thailand
TEXTURE CLASSIFICATION BASED ON SPATIAL DEPENDENCE FEATURES USING CO-OCCURRENCE MATRICES
AND MARKOV RANDOM FIELDS ...... .................. .239
WilLiam Schwartz and Hdlio Pedrini, Federal

SESSION MA-PZ: Image Filtering and Morphological Processing

DEFECT DETECnON ON HARDWOOD LOGS USING HIGH RESOLUTION THREE-DIMENSIONAL LASER SCAN DATA . . , 2 4 3
t i y a Thomas, Lamine Mili, Clifford A. Shaffeer; Virginia Polytechnic Institute and Srare University, USA; and Ed Thomas, USDA
Forest Senice Northeastern Researclt Station. USA
A ROBUST SPECKLE REDUCING ANISOTROPIC DIFFUSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Clovis Tuubeer;Hadj Butntia and Aloin Ayache, ENSEEIHT-IRIZ France
EDGE DETECTION BASED ON DECISION-LEVEL INFORMATION FUSION AND ITS APPLICATION IN
HYBRID IMAGE FILTERING .................................................. 251
Jia Li, Oakland University, of Posts and Telecommunications, China
ORDER FILTER WITH PROGRESSIVELY DECIMATED F'ILTERING WINDOW APPLICATION TO
COLOUR IMAGE ENHANCEMENT.. .................................. 255
Jocelyn Chanussor, Signal and Image Lnboratory, Saint Martin D Heres, France
ACCELERATING SVD ON RECONFIGURABLE HARDWARE FOR IMAGE DENOlSlNG .............................. .259
Aziz Ahmedsaid and Abbes Amira, Queens Universiry Belfasr, UK
IMAGE ENHANCEMENT USING STOCHASTIC RESONANCE . . . . . . . . . . . . . . . . . .263
Qinghua Ye, Haining Huung and Chunhua Zhang. Institute ofArous
FILTERING OF COLOR MAP IMAGES BY CONTEXT TREE MODELING 267
Pave1 Kopylov and Pasi Franri, University of Joensuu, Finland
TIKHONOV REGULARIZATION VERSUS SCALE SPACE A NEW RESULT 27 1
Luc Florack, Remco Duirs and loris Bierkens, Eindlroven University of Technology, The Netherlands
ESTIMATING THE PHASE CONGRUENCY OF LOCALISED FREQUENCIES . .
Peter Myerscougli and Mark Nixon, Southa,npron Universi?, UK
TEXTURE SEGMENTATION USING DIRECTIONAL EMPIRICAL MODE DECOMPOSITION.. . . . . . . . . . . . . . . . . . . . .279
Z1iong.ruon Liu, Hongjian Wang and Silong Peng. lnsritiite of Autoinorion, Chinease A
A MATCHING METHOD BASED ON MARKER-CONTROLLED WATERSHED SEGMENTATION . . . . .
yi Hu and 7imtnlmru Nrrgoo. Cradunte Sckool of Envimntnent und I~fivt,~atiun Sciencrr, Yokolm,no National Univer.sily, J c p ~ n
H-THINNING FOR GRAY-SCALEIMAGES'. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Srrbrina Rami-Slrojaei. Nest1e' resenrch Centec Schwizerlrmrl; and Corinne Mrcliiec CMLA, ENS Cachun, Fmnce

SESSION MA-P3: Image Enhancement I

MOTION-BASED SPATIAL-TEMPORAL IMAGE REPAIRING.. . . . ................................ . . . . . . .291
Wen-viZliao. SanioSf Corporation, USA
TEMPORAL FILTERING OF WAVELET-COMPRESSED MOTION IMAGERY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,295
Murk Ro6ertson. A FRL hzfu,rmtorion Dirrcro,nre/lFEC, USA
TWO LAYER SEGMENTATION FOR HANDLING PATHOLOGICAL MOTION IN DEGRADED POST PKODUCTION MEDIA . . .299
Benjamin Kent, Tlrr Found?, Englrmd; Ani1 Kokorom, Elecrmnic ond Electrical Engineering Dqm-tmenr, University of Dublin,
Trinity Collegc. Irelond; Llill Collir nnd Simon Robinson. Tlze Fotmdty England
OI'TIMAL BLOCK BOUNDARY PRE/POST-FILTERING FOR WAVELET-BASED IMAGE AND VIDEO COMPRESSION . . . . ,303
Jie Liang, Simon Frnser Uniiwsity, Canada; Clwzgjie Tu, Microsofr. USA; and Troc Tron. Jolins Hupkins liniversi?, USA
A FILTERING APPROACH TO EDGE PRESERVING MAP ESTIMATION OF IMAGES . . . . . . . .
~ Umid R~uhnan,University of NSW Airsimlirr
Dwid H u n ~ p l w cond
JOINT BLIND SEPARATION AND RESTORATION OF MIXED DEGRADED IMAGES FOR DOCUMENT ANALYSIS . . . . . . . 31 I
Anna Tonazzini. lslittrto di Scienzo e Tecnologie dell Infonnazione - Consiglio Nazionole delle Ricerche (lmly), Italy; / w i n Germe
and Francesco Cricco, Diparrimento di Mutematica e Infonnarica - Universitii degli Srirdi di Pengia, Italy
DESENSITISATION OF MEDICAL IMAGES RESTORATION UNDER CRUDE ESTIMATES O F MOBILE KADIO CHANNELS. . . 3 15
Gnillenno Cisneros, Pol?rechnic Universig of Madrid. Spoin; Emiliano Bemu 'es University of Znmgozo. Spain; Irma Rodriguez,
Miguel ngel Santiago and Fcderico Ivnrez, Polyteclmic Uniiwsity of Modrid, S p i n
ESTIMATING FIRST-ORDER FINITE-DIFFERENCE INFORMATION IN IMAGE RESTORATIONPROBLEMS 321
Patrick L. Combertes, Univcrsite Paris 6, France; and Jean-Cliristoplie Pesquer, Universite Murne la Vallee, France
RESTORATION OF HALFTONED COLOR-QUANTIZED IMAGES USING PROJECTION ONTO CONVEX SETS.. . . . . . . . . . 325
vik-Hing Fung rrnd Yuk-Hee Clurn, The Hong Kong Polytechnic University, Hong Kong
PERCEPTUAL IMAGE QUALITY ASSESSMENT BASED ON BAYESIAN NETWORKS..
Ronaldo Znmpolo and Rui Seora, Federal University of Santa Catarina, Brazil
KECOGNITION OF PARTLY OCCLUDED PERSON ACTIONS IN MEETING SCENARIOS. . . . . . . .
Martin Zobl, Andreas Lnikn. Frank Walllioff and Cerhurd Rigoll, Munich Universify of Teclinology, Instil
communication. Germany

SESSION MA-P4: Video Segmentation

ADAPTIVE SEGMENTATION FOR GYMNASTIC EXERCISES BASED ON CHANGE DETECTION OVER
MULTIRESOLUTIONCOMBINED DIFFERENCES ................................................................ 337
Josi M. Cobo, Luis Salgado and Julia" Cabrera, Universidad P o l i t h i m de Madrid, Spain
A KNOWLEDGE-BASED APPROACH TO DOMAIN-SPECIFIC COMPRESSED VIDEO ANALYSIS .................. ,341
Vasileios Mezaris, loannis Komparsiaris and Michael Strintzis, Informatics and Telematics Institute, Centrefor Research and
Technology Hellas, Greece
ANNOYANCE OF SPATO-TEMPORAL ARTIFACTS IN SEGMENTATION QUALITY ASSESSMENT
EIisa Drelie Celasca, Touradj Ebrahimi, EPFL - Swiss Federal Instilure of Technology, Switzerland: Mylene
and Sanjir Mitra, Univeririry of Calijomiu Santa Barbara, USA
HUMAN DETECTION IN GROUPS USING AFAST MEAN SHIFT PROCEDURE.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .349
Csaba Beleznai, Advanced Computer Vision CmbH - ACY Austria; Bemhard Frlihtiick, Siemens AC Austria, Austria; and Horst
Bischoj Inrtitu/e for Computer Graphics and Vision, Univ. of Technology Cruz, Austria
A PROBABILISTIC FRAMEWORK FOR SEGMENTATION AND TRACKING OF MULTIPLE NON RIGID
OBJECTS FOR VIDEO SURVEILLANCE. .............. ............................................. 3s3
Aleksandar Ivanovic and Thomas S. H u n g . ience and Technology, University of Illinois, USA
TEMPORAL STABILIZATION OF VIDEO OBJECT SEGMENTATION FOR 3D-TV APPLICATIONS ... ...... .357
Cigdem Eroglu Erdem, Momentum A.S., Turkey; Fabian Emst, Andre Rederr. Philips Research Laboratories, The Netherlands; and
Emile Hendriks, Delft Universify of Technology, The Netherlands
INTEGRATION OF MOTION AND IMAGE FEATURES FOR AUTOMATIC VIDEO OBJECT SEGMENTATION.. . . . . . .361
Wei Wei and Ngan King Ngi, Department of Electronic Engineering, The Chinese University of Hong Kong, China

vi
MOTION ESTIMATION AND DETECTION OF COMPLEX OBJECT BY ANALYZING RESAMPLED MOVEMENTS OF PARTS.. 365
Punpiri Piamso-ngo, Kaset.~nrrUniversiq, Tlmiland; and Nuboni Bnbagucki, Oxaka U n i r e r x i ~Japnn ,
MULTIPLE MOTION SEGMENTATION WITH LEVEL SETS WITHOUT PRIOR INFORMATION . . . . . . . . . . . . . . . . . -369
Ova Beshes, Cenrre d'Errrdes er de Reclrerclre des T~l~~oirsiliinicarions, Tuni.ria: and Ziad Belliadj Ecole SupL'rieure des
Coitenunicurions de Tunis, Tunisia
LV CONTOURTRACKING IN MRI SEQUENCES BASED ON THE GENERALIZED FUZZY GVF ....................... .173
Wufan Chen, Slioujvn Zlrou, Key I d J i i r Medico1 Image Processing. Depr. of B M 6 rlie Nrcr Milirory Medico1 Unii~.,CIrina; and Bin
Liang. Guang Dong Bmnclz of NSBC. China
FAST MOTION ESTIMATION AND MOTION SEGMENTATION USING MULTI-SCALE APPROACH 371
Cedric Dernonceoru, CREA/UMFA, France: and Djemorr Kachi-Akkoechc, CREA, Fmnce
A MOTION FIELD RECONSTRUCTION SCHEME FOR SMOOTH BOUNDARY VIDEO OBJECT SEGMENTATION.. . . . . . . ,381
Jean Gao, Ninrrrl Tlrakoor and Sungvong Jung. Univrrsig of Texni, USA
TEMPORAL VIDEO SEGMENTATION USING GLOBAL MOTION ESTIMATION AND DISCRETE CURVE EVOLUTION . . . , 3 8 5
Siripong Trrrrrrsnnaravom. Joerg Hcrrec Uw'c Rnusclwnbacl~,Siemens AG CT IC 2, Gennrmp: Andri Karp, Urriversip of Erlangem
Nure,,zherg. Clmir o f M d r i m d i r r Cr,,,i,,mnicorir,n.s cmrl Signal Processing, G e n ~ m y ~; n Klaus
d Illgner; Siameni AG CT IC 2, Geniiany

SESSION MA-PS: Low-level Image Indexing and Retrieval

USING MULTISCALE TOP POINTS IN IMAGE MATCHING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .189
Brmi Plrirel, Luc Flo,-ock, F r m s Kmrerr ond Evg~wniaBabnacAno~n,Technischc Univemireir Eiridlmwz, TIE Netlzrrlands
ROTATION INVARIANTTEXTURE FEATURES USING ROTATED COMPLEX WAVELET FOR CONTENT
BASEDIMACERETRIEVAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 ~ 3
Mancsh Kokarr, P K. Bis,wrs and B. N. Cliotterji, Indicm lnsriture <fTeclinolog? Klmmgpur. India
A FRAMEWORK FOR SOFT HASHING AND ITS APPLICATION TO ROBUST IMAGE HASHING. . . . . . . . . . . . . . . . . . . . . . . ,397
Eliznherk McCarrlty, Felir Balodo, Gucnole Silvestrr and Neil H u r l e ~Univeniry College Dublin, Irdnnd
COMBINING LOCAL CLASS PAlTERNS AND DISCOVERED SEMANTICS FOR IMAGE RETRIEVAL . . . . . . . . . . . . . . . . . . .401
Joo-Hww Lint, lnsrirure for Inficomm Research, Singapore; and Jesse Jin, Universiry of New Sour11 W~des,Amrralia
IMAGE RETRIEVAL BASED ON PROJECTIVE INVARIANCE . . . . . . . . . . . . . . . . ....................... ,405
Sirblznsis Clznudlrrrri, Vinay P Namboodiri, /IT Mumbai, India: and Rajaslieklur S., I1
A NOVEL 2D SHAPE MATCHING ALGORITHM BASED ON B-SPLINE MODELING.. ................................ ,409
Wung Yue ond Troh Eam Khwang, Nanyang Technological University, Singupore
CVPlC IMAGE RETRIEVAL BASED ON BLOCK COLOUR CO-OCCURANCE MATRIX AND PAlTERN HISTOGRAM ..... .413
Gerald Schaefer, Simon Lieutaud. Noaingliaiii Trent Universin: UK; and Cuoping Qiu, University of Norringhain, UK
SYMMETRY FEATURE IN CONTENT-BASED IMAGE RETRIEVAL ................................................ .417
Jingrui He, Eingliua Universiry, China; Mingjing Li, Hongjinng Zhang, Microsoft Research Asia, China; and Changslrui Zlrang,
Tsinghua University, China
REGION CORRESPONDENCE FOR IMAGE RETRIEVAL USING GRAPH-THEORETIC APPROACH AND
MAXIMUM LIKELIHOOD ESTIMATION 42 I
Church-Yu Li and Chiou-Ting Hsu, National
EFFICIENT IMAGE CODING FOR ACCESS TO PIXEL RANGES ................................................... .425
Sehoon Yea, Rensselaer Polytechnic Insritute, USA: Amir Said, Hewlett Pucknrd Laboratories, USA; and William Pearlman, Rensselaer
Polytechnic Insrirute. USA
COMPRESSED DOMAIN FEATURE TRANSFORMATION USING EVOLUTIONARY STRATEGIES FOR
tlMAGECLASSlFlCATlON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429
Chun l p Chiu, Hau San Wong and Horace H S lp, Department of Computer Science, City University ofHong Kong. Hong Kong
IMAGE DATABASE RETRIEVAL USING SKETCHED QUERIES ................................................... ,433
Abdolah Chalechale, Golshah Naghdy and Prashan Premaratne, University of Wollongong. Australia
JPEC~OOO v s . JPEG FROM AN IMAGE RETRIEVAL POINT OF VIEW 437
Gerald Schaefer: Norringham Trent University, UK

SESSION MA-P6: DCT-based Video Coding

A VIDEO CODING SYSTEM FOR SIGN LANGUAGE COMMUNICATION AT LOW BIT RATES 441
Dimirris Agraflotis, Nishan Canogarojah, Dave Bull, Jim Kyle, Helen Seers, University of Bristol, UK; and Matthew Dye, Universiry
of Rochesrec USA
DCT-BASED PHASE CORRELATION MOTION ESTIMATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,445
Min Li, Mainak Biswas, Sanjeev Kumar and Truong Nguyen. UCSD. ECE Dept., USA
A NEW RATE CONTROL SCHEME FOR H.264 VIDEO CODING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,449
Peng Yin and Jill Boyce, Corporate Research. Thomson Inr, USA

vii
AN EARLY DETECTION OF ALL-ZERO DCT BLOCKS IN H.264 .......... . . . . . . . . . . -45s
G p Yeong Kim, Department of Electronics Engineering, Pusan Nafional Universiy, Soutli Korea; Yong Ho Moon, Divi.sion of
Digital and blfonnarion Engineering. Pusnn University of Foreign Sredies, Sourlr Ko,ra: and Jae Ho Kim, Depnrt,nenr of
Electronics Engineering, Puson Nafional Universiy, South Korea
ENERGY SCALABILITY IN MULTIMEDIA CODEC USING NQDCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .457
Ming-Yan Clran and Chi- Wah Kok, Depr. EEE, Hong Kong Universiy of Science and 7 ~ c I m o l o gHong ~ Kong
A FAST H.264 INTRA PREDICTION ALGORITHM USING MACROBLOCK PROPERTIES 46 I
Cliun-ling Yang, College of Elcctcrmnic and Informarion Engineering, South Clinn Uniwemiy of
Wing-Hung Lmn, Depamnmt of Electronic Engineering, C i q Universiq of Hong Kong. Hong Kong
INTER FRAME CODING WITH TEMPLATE MATCHING SPATIO-TEMPORALPREDICTION. .......................... ,465
Kazuo Siigi~iioto,Sadaatsu Kato, Clioony Seng Boon. Mirsuru Kobayishi and Yoshinori Suzuki, NTT DoCoMo. Inc., Jq,nn
NEW SCALING TECHNIQUE FOR DIRECT MODE CODING IN B PICTURES. ................. . . . .469
Xiangyang Ji, Debin Zlruo, Wen Goo, lnsritutr of Computing Teclrnolog)~.Cliinessr A nces, Beijing, China; Harlin
1ii.vtitiite of Technology,' Harhin, Clrina: and Siwei Ma,lnsrirure of Cmnprrting Technology, Cliinesc Acade,rzv cif Sciences, Beijing, Chino
TREE STRUCTURED HYBRID INTRA PREDICTION.. . . . . . . . . . . . . ................... . . . . . . . . . . .47s
Sato,slri Kondo, Hisrro S a a i and Slrinya Kadono, Matsui1,itrr Electric hdrtsrrial Co.. 1,td. Jqmn
PERCEPTUALLY WEIGHTED DISTORTION MEASURE IN LOW BITRATE BLOCK BASED VIDEO CODERS . . . . . . . . . . . . ,477
.Iaef?unIn, Ali Jerbi, UU Video Inc., Canada: and Foulred Ben Amara, Uniwvsit? of Tor<Jllto,Canada
VARIABLE BLOCK-SIZE TRANSFORM AND ENTROPY CODING ATTHE ENHANCEMENT LAYER OF F G S . . . . . . . . . . . . .481
Jungong Hun, ISN national Iahorato,~ofXiliim Univers o y n SIIII, Feng Wu, Slripng Li, Microsoft Research Asia, China;
ond Zliaoyong Lu, ISN nrrtional /ahoratory of Xidian Uni
HIERARCHICAL MODE SEARCH WITH CLASSIFICATION OF BISECTIONAL PREDICTION MODES BASED ON
THE POSITION OF MOTION BOUNDARY.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 4 8 5
Sadoafsrt Karo, Kazeo Sugbnoto, Fulvio Mosclrefti and Clroon Seng Boon, N7T DoCuMo, Inc., Japan
PERCEPTUALLY-ADAPTIVEPRE-PROCESSING FOR MOTION-COMPENSATED RESIDUE IN VIDEO CODING . . . . . . . . . ,489
Xiaokanng Yong. Wei.si Lin, Zlzongkanng Lu, Er Ping Ong and Susa Yao, lnsrirure for Infoconm Researclr, Singapore

SESSION MA-W: Image Compression and Applications

JPEG-MATCHED DATA FILLING OF SPARSE IMAGES.. . . ............. . . . . . . . . . . .493
George Pavlidis. Sola Tsekeridou and Chrirrodoulos Chamzos, of Thrace, Cree
A WO-DIMENSIONAL LIRING SCHEME OF INTEGER WAVELET TRANSFORM FOR LOSSLESS IMAGE COMPRESSION. ,497
Yan-Kui Sun, Tsinghua University, Clrinu
PALE'ITE REORDERING UNDER AN EXPONENTIAL POWER DISTRIBUTION MODELOF PREDICTION RESIDUALS. . . .501
Arrnando Pinho, DET/lEETA, Universiry of Aveiro, Portugal; and Anr6nio News. IEETA. University ofAveim, Portugal
A COMPARISON OF NON-ORTHOGONAL AND ORTHOGONAL FRACTAL DECODING .......... .SO5
Ming Hong Pi, Anup Basu. Dept. of Computing Science, University of Alberta, Canada: Mrinal
Computer Engineering. University of Alberta. Canada: and Hua Li. Dept. of Mathematics and Computer Science. University of
Lethbridge, Canada
ANALYSIS/SYNTHESIS SYSTEMS FOR PROGRESSIVE-TO-LOSSLESS EMBEDDED WAVELET IMAGE CODING 509
Kunitoshi Kornnrsu and Kaoru Sezaki. University of Tokyo, Japan
COMPARISON OF LOSSY TO LOSSLESS COMPRESSION TECHNIQUES FOR DIGITAL CINEMA . . . . . . . . . . ,513
Stefano Andriani, Giancarlo Calvagno, Tomaso Eiseghe, Gian Antonio Mian, DEI - University of Podova, Italy; Marc0 Durigon,
Roberto Rinaldo, DIEGM - University of Udine. Iraly; Mike Knee, Paul Walland, Snell& Wilcox Lrd.. UK; and Michael Koppetz,
Arnold & Richter Cine Technik (ARRIJ, Germany
LOSSLESS COMPRESSION OF COLOR MOSAIC IMAGES .......................... ,517
Ning Zhnng and Xiaolin Wu, McMasrer Universiy, Canada
DEMOSAICKING AND JPEGZOOO COMPRESSION OF MICROSCOPY IMAGES ..................... 521
Beno"if Parrein, Marc Tarin and Patrick Horain, GETNNI: France
QUALITY AND COMPLEXITY COMPARISON OF H.264 INTRA MODE WITH JPEG2000 AND JPEG .......... ,525
Aravind AI. Bindu II Roo, Sudhir S.Kudva, National Institute of Technology Karnaroka, Surarhknl, India:
Instruments (India) Ltd., Bangalore, India: Sumam David S., National Institute of Technology Kamarako. Surarhlwl, India; and Ajit
V Rao, Texas Instruments (India) Lfd., Bangalore, India

SESSION MA-PS: Distributed Source Coding and Others

DISTRIBUTED COMPRESSION OF THE PLENOPTlC FUNCTION. ................................ . . . . . . . . . . .S29
Nicolas Gehrig and Pier Luigi Dragorti, Imperial College London, U K
JOINT DATA COMPRESSION AND ERROR PROTECTION FOR COLLABORATIVE TRANSMISSION . . . . . . . . . . . . . . . . . . . .533
Hairong Sun, Mihaeln Van der Schaar and Zhi Ding, Universiry of Califomia Davis, USA
ERROR RESILIENCE SUPPORTING BI-DIRECTIONAL FRAME RECOVERY FOR VIDEO STREAMING. . . . . . . . . . . . . . . . . ,537
Kai-Clmu Yring. Chun-Ming Huang and Jia-Slrrmg Wang, Depurrmmr of Cornpurer Science, Tying Hun Univwsity, Taiwan
H.264BASED STREAM MORPHING WITH SCALABLE MOTION CODING 54 I
James Maniicol, Micltael Fmrer and Jolm Amold, University College, The University of New Suurh Wales, Austmliu
H.264/AVC DATA PARTITIONING FOR MOBILE VIDEO COMMUNICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,545
Thomas Srocklinnme,: Munich Universiq of Technology. German>'; and Mojo Bwrrom, Bosron University, USA

SESSION MP-S I : Drlormable Models and Applications

APPROXIMATION OF IMAGES BY BASIS FUNCTIONS FOR MULTIPLE REGION SEGMENTATION WITH LEVEL SETS.. ,549
Cnrlos V ~ Z ~ L IINRS-EMZ
L'Z. Canada; Abdol-Rezrr Munsuuri, DEAS, Honm-d Univer.sity, USA: and Airiar Miricke, INRS-EMZ Canada
JOINT DENSE 3D INTERPRETATION AND MULTIPLE MOTION SEGMENTATION OFTEMPORAL IMAGE SEQUENCES:
A VARIATIONAL FRAMEWORK WITH ACTIVE CURVE EVOLUTION AND LEVEL SETS. 553
Hiclaen ScLkari and Amor Mificlie, INRS-EMZ Canudu
ACTIVE CONTOUR MODELS - A MULTISCALE IMPLEMENTATION FOR ANATOMICAL FEATURE DELINEATION
IN CERVICAL IMAGES . . . . ................................................................ 557
Viuru Vun Road, The Universi
PERSON AUTHENTICATION USING ASM BASED LIP SHAPE AND INTENSITY INFORMATION 61
Lin L~rrngM o t and Wing Hong l a , , Cir? llnivwsir? of Hong K o u ~ Hong
. Kong

SESSION MP-S2: Media Security Issues in Streaming and Mobile Applications

NETWORK FRIENDLY MEDIA SECURITY: RATIONALES. SOLUTIONS, AND OPEN ISSUES .................... 565
Wenjun Zeng, Xinluin Zliimng arid Junyiang Iun, Uni\,. ~~Missuun'-CoIrr,nbio,
USA
SECURITY EVALUATION FOR COMMUNICATION-FRIENDLYENCRYFTION OF MULTIMEDIA ...................... ,569
Yiniiin Mrro rind Min Wu, ECE Deparmenr, liniversity of Mu&md, College Purk. USA
A NOVEL LOSSY-TO-LOSSLESSWATERMARKING SCHEME FOR JPEG2000 IMAGES ...................... ,573
Zltishou Zlmng. Qibin Sun and Wai-Ckoong Wong, lnsrirure for Infucmnm Research, Singupore
SECURE TRANSCODING WITH JPSEC CONFIDENTIALITY AND AUTHENTICATION ............................... .577
Susie Wee and John Aporrolopoulos. HP Lobs, USA

SESSION MP-L1: Face Detection, Recognition, and Classification I

ESTIMATING THE QUALITY OF FACE LOCALIZATION FOR FACE VERIFICATION ............................... 581
Yann Rodriguez. Fabien Cardinnu, Sarny Bengio and Johnny Marikthuz. IDIAI: Switzerlo
STATISTICAL TRANSFORMATIONS OF FRONTAL MODELS FOR NON-FRONTAL FACE VERIFICATION. 585
Conrad Sanderson. Universiq of Adelaide, Australia: and Samy Bengio. IDlAP Researclr hstitule, Swirzerland
EFFICIENT FACE ORIENTATION DISCRIMINATION. ............................................................. ,589
Shumeet Baluja, Mehran Sahnnzi and Henry Rowley, Google, USA
OMNI-DIRECTIONAL FACE DETECTION BASED ON REAL ADABOOST. . . . . 593
Chnng Huang, Bo Wu. Deparrmenr of Computer Science and Techno1ug.y. Tsinghua Universiry, China; Haizhou Ai, Department of
Computer Science and Technology, Tsinghua University. Beijing IW084, PR.Clrinn, China; and Slrihong Lao, Sensing Technology
Lahurarory, Oinmn Corporation. Japan
A ROBUST FACE DETECTOR UNDER PARTIAL OCCLUSION.. . . . . . . . . . . . . . . . . . . .
Kazuhiro Horra, The University of Elecrm-Cunimunicatiuns, Japan
MEAN-SHIFT BASED MIXTURE MODEL FOR FACE DETECTION IN COLOR IMAGE
Tze-fin Chow and Kin-Man Lam, The Hung Kong Polytechnic University, Hung Kong
STEREO HEAD/FACE DETECTION AND TRACKING .............................................................. -605
Jian Gang Wang, Ronda Venkatesworlu and Eng Thiarn Litn, lnsrirure fur Infucumm Research, Singapore
MARGIN-MAXIMIZATIONDISCRIMINANT ANALYSIS FOR FACE RECOGNITION ............................... 609
Yan Zru and Eric Sung, Nanyang Technological Universiq, Singapore

SESSION MP-L2:.Video Summarization and Browsing

A RATE-CONSTRAINED KEY-FRAME EXTRACTION SCHEME FOR CHANNEL-AWARE VIDEO STREAMING . . . . . . . . . . ,613
Yu-Hsuan Hu,Nariunal Chung Clieng Universio, Taiwan; Wei-Ren Chen, UleNend, Taiwan; and Chk-Wen Lin, National Chung Cheng
Univet-siry, Taiwan
OFTIMAL VIDEO SUMMARIZATIONWITH A BIT BUDGET CONSTRAINT. ........................................ ,617
Zhu Li, Murorola Labs, Schaumburg, USA; Guido Schusrez Hochschule for Technik Rapperswil. Swirzerland; Aggelus Karsaggelus,
Northwestem University, Evansron, USA; and Bhavan Gandhi, Motorola Lobs. Schaumburg, USA
AN IMPORTANCE MEASUREMENT FOR VIDEO AND ITS APPLICATION TO TV NEWS ITEMS DISTILLATION. . . . . . . . . ,621
Jin-Hau Kuo, Chin-Wei Fang. Jen-Ha0 Yeh and Ja-Ling Wu. Communication and Mulrimedio Lahorarory, Depnmnenr of Compurer
Science and Infomarion Engineering, National Taiwan University, Taiwan

ix
A NEW APPROCH TO AUTOMATIC MUSIC VIDEO SUMMARIZATION.. . . . . ................................... ,625
Xi Slrao, Cliongsheng Xu, Institute for Infocmnm Research, Singapore; and Mohan Kankonnlralli. Narionol Uniwrosity of Singapore,
Singapore
A GENERIC MID-LEVEL REPRESENTATION FOR SEMANTIC VIDEO ANALYSIS. . 629
Qing Tung, School oflnfonnariun Teclmologies, Univer.yirv of Sydney, Ausrrdio; Joo-Hwee
Singapore; Jesse S Jin, S d ~ o oolf hfoonnorion Technologies, University @“Sydney,Ausrrolia: Hniping Sun, hzrtiture for InJo~oinnr
Res~orrlr,National Uniuersitg of Singupore, Singapore; and @iTim. Insrirurefor Infi,cotnin Re.search, Singupow
DYNAMIC BAYESIAN NETWORK BASED EVENT DETECTION FOR SOCCER HIGHLIGHTEXTRACTION.. . . . . . . . . . . . , 6 3 3
Fe; Wang, hutirrire of Compuring Teclinolog~,Chinese Acadany of Sciences, China; Yu-Fei Ma, Hong-Jiang Zlrang, MicrosoJr Resrorch
Asiu. Chino; and Jin-Too Li, Instirrue of Co,nputing Technolog,: Clrinese Acadetnp ofScimces, Chino
GROUPING VIDEO SHOTS INTO SCENES BASED ON ID MOSAIC DESCRIPTORS.. . ..... 637
Henri Nicolo.~,Anne Manoury, INRIA. France; Jenny Benois-Pinert~r.Lnbri. University of Bordeour. France; Willio,n Dupuis and
Dominique Burbu, IRCCyn. France
DISTRIBUTEDMPEG-7 IMAGE INDEXING USING SMALL WORLD USER AGENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .641
PanagiorisAnrlroursor, Amadeh Kurliki, Universirv of Toronto, Canada; Dimirrios Anrlmnl.so.v, Ryerson P~lyrrchnicUniver.siry
Canada; Kosranrinos Plaronioris ond Annsrusios Venrrinno~Joiil~~s.
Universiq of Toronto. Cunado

SESSION MP-L3: lniage Filtering and Partial Diffcrcntial Equations

AUTOMATIC IMAGE DECOMWSITION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,645
Kedur Par~rtwardl,a,rond Gaillermo Sripiro. Depr. oJ‘ Electrical Ehgineering ond Digiral Trclmology Cenres Unierrsiry of Minnesota.
Minneapo1i.r. USA
A PARTIAL DIFFERENTIAL EQUATION APPROACH TO IMAGE ZOOM. ........... ,649
Ceminade, France; and Frederic Guichard. Do Labs,
Abdulriroenint Be1~1~z111idi.
A NEW APPROACH TO THINNING BASED ON TIME-REVERSED HEAT CONDUCTION MODEL.. . . . . . . . . . . . . . . .6S3
Xinhua J i and JuJr, Fmg, School of Electronics Engineering and Coinputer Science, Peking Clnirursity, Cliinii
MULTIPLICATIVE MULTIRESOLUTION DECOMPOSITION FOR 2-D SIGNALS: APPLICATION TO SPECKLE
REDUCTION IN SARIMACES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6S7
Serir Amino, U.S.TH.B. Algeria; ond Belouchrani Adel, E.N.F: Algeria
ISOTROPIC-POLYHARMONIC B-SPLINES AND WAVELETS. . . . . . . . . . . . . . . . .........................
Dirnirri Van De Ville, Thierry Blu, Brigirre Forsrer and Michael Unses Swiss Federal lnsrirure of Technology Dlusunne, Switzerland
AN IMPLEMENTED ARCHITECTURE OF DEBLOCKING FILTER FOR H.2MIAVC ................................... ,665
Bin Slieng, Departmenr of Coinpurer Science and Technology,Harbin lnsrirure of Technology, China, China; Wen Cao, lnsrirure of
Computing Technolog): Chinese Academy ofkiences, Cliina, China; and Di Wu.Deparrmenr of CoJnpurer Science and Technology,
Hnrbin Insrirure of Technology, China, China
A NEW SIMILARITY MEASURE USING HAUSDORFF DISTANCE M A P . . ................................ . .669
Etienne Baudriez Cilles Millon, Fr’ed’eric Nicolier and Su Ruan. CReSTIC, France
CONTOUR SIMPLIFICATION USING NON-LINEAR DIFFUSION ................. .......... ,673
Antonio Pinheim, Universidade da Beira Interioz Porrugal; and Mohrnmed Ghnnbari, Un

SESSION MP-L4 ImageNideo Indexing and Retrieval

ROBUST PERCEPTUAL IMAGE HASHING USING FEATURE POINTS.. . . . . . . ........... ,677
Vishal Monga and Brian L. Evans, Universiry of Texos, Austin. USA
FEATURE STATISTICAL RETRIEVAL APPLIED TO CONTEN-BASED COPY IDENTIFICATION. . . . . . . . . . . . . . . . . . . .
Alexis Joly, Olivier Buisson, Instirut National de I’Audiovisuel (INA), France; and Carl Fr’elicor. Universit4 de La Rochelle. France
ROBUST VIDEO SIGNATURE BASED ON ORDINAL MEASURE .............................. 685
Xian-Sheng Hua, Microsof! Research Asia, China; Xian Chen, Tsingh Cliina: and HongJiong Ulang. Micros
Researrh Asia. China
IMAGE RETRIEVAL BASED ON FEATURE WEIGHTING AND RELEVANCE FEEDBACK. . . . . . . . . . . . . . . . . . .
Mohummed Lamine Kherfi and Djemel Ziou, D ’eparremenr d’infomarique, Universir’e de Sherbmoke, Canada
SEMANTIC-BASED TRAFFIC VIDEO RETRIEVAL USING ACTIVITY PA’ITERN ANALYSIS. ..............
Dan Xie, Weiming Hu, Tieniu Tan, National Laborarory oJPartem Recognirion, China; and Junyi Peng, Beijing Uni
Aemnaurics and Asrmnaurics, China
DETECTION OF UNIQUE PEOPLE IN NEWS PROGRAMS USING MULTIMODAL SHOT CLUSTERING.. ............... , 6 9 7
Cwteyr Tuskiran, Purdue Universiry, USA: Alberro Albiol. Technical University of Valencia. Spain: Luis Torres. Technical Universiry
of Catalonia, Spain; and Edward Delp, Purdue Universiry, USA
A SIMPLE AND FAST COLOR-BASED HUMAN FACE DETEmION SCHEME FOR CONTENT-BASED
INDEXING AND RETRIEVAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sangkeun Lee and Monson H. Hayes, Georgia lnsrirule oJTechnolog.v, USA
A PINHOLE CAMERA MODELING OF MOTION VECTOR FIELD FOR TENNIS VIDEO ANALYSIS. . . . . . . . . . . . . . ,705
Peng Wang. Rui Cui, Bin Li and W i - Q i m g Yang, Depr. Cornputer Science, Tsinghua Univ., Beijing, Cliina

X
SESSION MP-LS: Watermarking 11
BLIND SOURCECAMERAIDENTIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7~
Melidi Kl~arrrrzi,Hasrev 7: Sencar arid Nosir Menion, Pol?technic UniversitJ, USA
BI-LEVEL IMAGE WATERMARKING FOR IMAGE AUTHENTICATION SURVIVING JPEG LOSSY COMPRESSION 713
Jagdi.sh C. Palm, Clmng How Em and Ee-l,uang Ang, Nanyaiig Technological UniversitJ, Singqmre
QIMWATERMARKLNGGAMES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 717
Ani1 Kunmr Goteri and Pierre Moulin, Univrrsiry of lllinoi.~.~ r b r l n a - ~ ~ l u l l l ~ J oUSA
ign,
IMPROVE SECURITY OF FRAGILE WATERMARKING VIA PAKAMETERIZED WAVELET 72 I
Jiwu Hwmg, Jirnqunn HLI.Daren H u n g , Sun Ynt-Sen Uniwrsip, China; rind fim Q. Slri, New Jer
A ROBUST SPREAII-SPECTRUM WATEKMAKKING METHOD USING TWO-LEVEL QUANTIZATION.. . . . . . . . . . . . . . . . . ,725
Antirony TS. Ho ond Feng Shu, Nnnyong E~clmulogicalUniversity, Singapore
WATERMARKING 71) SHAPES USING LOCAL MOMENTS 729
Adrian I3or.c Univcrsily cf York. UK
AN ATTACK TO HPCS-STEGANOGRAPHYUSING COMPLEXITY HISTOGRAM AND COUNTERMEASURE . . . . . . . . . . . ..713
Mickilioru Niimi, Tomolrito Ei, Hidrki Nudo, Eiji Kuunguchi, Kyrsliu /lJSlillrle of Technolog): J U ~ U I Jarid
; Bruce Segee. Universit?.
~ f M a i n u ,USA
A RST RESILIENT OBJECT-BASED VIDEO WATERMARKING SCHEME. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .737
Dc+zn He and Qibiri S J S ~Instirurc~Jbr
, Infucorrm Research (IZR), Singapore

SESSION MP-PI: Video Compression Standards I1

FAST MACROBLOCK MODE SELECTION BASED ON MOTION CONTENT CLASSIFICATION IN H.264lAVC . . . . . . . . . . . .741
Ming Liri,? nndWenskeng Wang, Dept. of Electronic Engineering, Tsinglrurr Uni\r. China
ADAPTIVE RATE CONTROL FOR H.264 ... 74s
Zvnggrco Li, Feng Pan, Keng Pang Lim. Xiao Lin and Susanro Rahm-dja, lnsrirrrtc for Infocomnt Rwmrck, Singopore
TIME-EFFICIENT LEARNING THEORETIC ALGORITHMS FOR H.2W MODE SELECTION. . . . . . . . . . . . . . . . . . . . . . . . .749
Arhisk Jqmukrm, UniversitJ of Illinois, Urbuna-Clmiitl~uign,USA: and Krishna Rorakonda, IRM 7:J. Wa Research Cenrre, USA
RATE-DISTORTION OPTIMIZED VIDEO CODING WITH STOPPING RULES: QUALITY AND COMPLEXITY . . . . . . . . . . . . ,753
Marcus Moecke and Rui Seam, Federul Universiry of Santa Caturina, Brazil
REAL-TIME H.264IAVC CODEC ON INTEL ARCHITECTUR
Vaughn Iverson, JeffMcVeigh and Bob Reese. Intel Corporation.
OPTIMAL BIT ALLOCATION FOR MPEG-4 MULTIPLE VID
Zltmzhong Chen and King Ngi Ngan, Deparrmenr of Elecrmnic Engineering, The Chinese Univcrsiry of Hung Kong. Hong Kong
LOW-COMPLEXITY MACROBLOCK MODE SELECTION F
Hyungioon Kim and Yucel Altunbasak, Georgia Imtitule of Technology. USA
FEATURE-BASED INTRA-PREDICTION MODE DECISION FOR H.264 ............................................. .769
Changsung Kim, University of Southern California, USA; Hsuan-Huei Shih, Ali Microelectronics Corp., USA: and C. -C. Jay Kuo,
University of Southem Califomia, USA
COEFFICIEKT TIIRESIIOI.DING ASD OPTIMIZEI) SELECTION OF TI IE LAGRANGIAN LlIII.TIPLIER
bOR NOS-REFERENCE FKAhlES IK l1.?01 V11)EOCOI)ING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 773
Ponrus Curlmm, Linkopings Universitet, Sweden; E Pan, Institute for Infocomm Research, Singapore; and Liang- 7ien Chia,
Nanyang Technological Univeniry, Singapore
EFF'ICIENT MEMORY MANAGEMENT CONTROL FOR H.264 777
Hyukjune Chung and Antonio Ortega, Universily of Southem Califomia, USA
FAST INTRA MODE DECIXON ALGORITHM FOR H.264lAVC VIDEO CODING ..................................... ,781
E Pun, X. Lin, S.Rahardja, K. Lim, Z. Li, Institute for lnfocomm Research, Singapore; C. Zhu. NTU, Singapore: D. Wu. S.Wu,
Institute for lnfocomm Research, Singapore: M! Ye and Z. Liung, NTU, Singapore
INTER-PLANE PREDICTION FOR RGB VIDEO CODING ......................................................... ,785
Woo-ShikKim, Dae-Sung Clio and Hyun Mun Kim, Samsung Advanced Institute of rechnologx South Korea
FAST MACROBLOCK INTER MODE DECISION AND MOTION ESTIMATION FOR H.264lMPEG-4 AVC 789
Zhi Zhou and Ming-ling Sun,Department of Elecrricol Engineering, Universiry of Washington,USA
A STUDY ON FAST RATE-DISTORTION OPTIMIZED CODING MODE DECISION FOR H.264.. ........................ ,793
Akiyuki Tanizawa. Shinichiro Koto, Tukeshi Chujoh. Toshiba Corporation, Corporate Research & Development Cenre,: Multimedia
LnborutoT, Japan: and Yoshihiro Kikuchi, TOSHIBA Corporation, Core Technology Cenrez Japan

SESSION MP-PZ: Error ResiliencdConcealment I1

MOTION-BASED SHAPE ERROR CONCEALMENT FOR OBJECT-BASED VIDEO . . . . . . . . . . . . . 797
Luis Soares and Femundo Pereira, Institutu Superior Tdcnico/ lnstituto de Telecornunicacoes,Portugal

xi
RATE DISTORTION ANALYSIS OF LEAKY PREDICTION LAYERED VIDEO CODING USING QUANTIZATION
NOISE MODELING.. . ....................................................................... 80 I
Yuxin Liu, Purdrte Universip, USA; Jusep Prade.s-Nebot, Univerridad Politecnica de Valencia, Spain; Pard Solmrm, Indinno
Univusirv-Pur& UniversiQ Indianapolis, USA; and Edward Delp, Purdue University, USA
MARKOV RANDOM FIELD ESTIMATION OF LOST DCT COEFFICIENTS IN JPEG DUE TO PACKET ERRORS . . . . . . . . . . .805
Jinwlio Yang a d Edward Delp. Ptirdue University, USA
EDGE DIRECTED FILTER BASED ERROR CONCEALMENT FOR WAVELET-BASED IMAGES.. . . . . . . . 809
Sliuiming Ye, @bin Sun, Institutefor Infoco,nm Researrh, Singapore; and Ee-Clrien Chang, National Universip ore
MINIMIZING A WEIGHTED ERROR CRITERION FOR SPATIAL ERROR CONCEALMENT OF MISSING IMAGE DATA.. . . ,813
Korrin Meisinger arid Andr ’e Kaup, Universip of Erlangen-Nuremberg, Germany
DESlNG OF REVERSIBLE VARIABLE-LENGTH CODES USING PROPERTIES OFTHE HUFFMAN CODE AND AVERAGE
LENGTH FUNCTION ...................... .................................. .E17
Wook-Hyun Jeong, Satii.nmg Elecrronics Co.. KO S7: Korea
ERROR RESILIENCE ANALYSIS OF MULTI-HYPOTHESIS MOTION COMPENSATED PREDICTION FOR VIDEO CODING
Wei-Ying Kung, Univer.it), of Southern Califumiu, USA; Clmng-Su Kiitt, tlie Cliinese Universip of Hong Kong. H ~ n gK o q ; and
C.-C. Jup Kim. U n i w w i h ofSouthern Cal$omia, USA
AN INTRA-FRAME ERROR CONCEALMENT: NON-LINEAR PATTERN ALIGNMENT AND DIRECTIONAL
INTERPOLATION ....................... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,825
Wiitripong Kuma.ilui.srhok, K i q Mongkut Univer.si 1;mnX H<wrimg, Eyicssoii Reseorclr Lnh, C;mrwny
EFFICIENT ERROR RECOVERY FOR MULTIPLE DESCRIPTION VIDEO CODING . . . . . . . . . . . . . . . . . 829
G~mJun Zhung and Rohert Stevenson, Univosit>~of Notre L h w , USA
EFFICIENT PACKET LOSS PROTECTION FOR JPEG2M)O IMAGES ENABLING HACKWARD COMPATIBILITY
WITH A STANDARD DECODER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,833
K l ~ a i n iMunadi,
l Masayrki Kwosu slii Kija, Tokyo Merropolitan Univemip, Jcipan
ERROR-RESILIENT WIRELESS VIDEO TRANSMISSION USING MOTION-BASED UNEQUAL ERROR
PROTECTION AND INTRA-FRAME PACKET INTERLEAVING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 837
@i Qu. Electrical and Cumpuler Engineering Dept., Univ. of Califomiu San Diego, USA; Yong Pei, Computer Science and
Dept., Wright Stole h i % ,USA; Jatncs W Modestino and Xuslreng 7bn. Electrical rind Corupuler Engineering Dept., Univ. ofMiami, USA
ON PACKETIZATION OF JPEG2000 CODE-STREAMS IN WIRELESS CHANNELS ......................... 841
Hua Cai, Microsoji Research Asia, China; and Sing Zeng, 7he Hong Kong University of Science and Technology, Hong Kong
CONTENT-BASED PERIODIC MACROBLOCK FOR ERROR-RESILIENTTRANSMISSION OF H.2M VIDEO.. . . . . . . . . . . . -845
Jinghong Zlieng and Lnp-Pui Cliau, Nanyang Teclinological Universit); Singapore

SESSION MP-P3: Biometrics I

CLASSIFICATION OF FINGERPRINTS USING SINGULAR POINTS AND THEIR PRINCIPAL AXES. . 849
Cluinarong Klitnaner and Dinh Thong Nguyen, University of Tasmania,Australia
A SVM-BASED METHOD FOR FACE RECOGNITION USING A WAVELET-PCA REPRESENTATION OF FACES.. ......... 853
Majid Safari, Mehrtasli Tafauoli Harandi and Babak Aruubi. Depamnenr of Elecrrical and Computer Engineering, University of Tehran, Iran
AN ANGULAR TRANSFORM OF GAIT SEQUENCES FOR GAIT ASSISTED RECOGNITION. .......................... ,857
Nikolaos Boulgouris, Konsrantinos Plaranioris and Dimitris Hatzinakos, University of Toronto. Canada
IRIS FEATURES EXTRACTION USING WAVELET PACKETS. . . . . . . . . . . . . . . . . . .
Erik bdgren, Thomas Ea, F r i d k r i c Amid, Florence Rossnnr and A m ” Amara. lnstirur
DISCRIMINANT IRIS FEATURE AND SUPPORT VECTOR MACHINES FOR IRIS RECOGNITION ...................... ,865
Byungjun Son, Hyunsuk Won, Gyundo Kee and Yillbyung Lee, Division of Cornpurer and Information Engineering. Yonsei University,
South Korea
NOISE REMOVAL AND IMPAlNTlNG MODEL FOR IRIS IMAGE. .................................................. 869
Junzhou Huang, Yunhong Wang, Jiali Cui and 7ieniu Tan. National hhomtory of Patfem Recognirion, lnsritute ofAuromation.CAS. China
ONE-TIME CALIBRATION EYE GAZE DETECTION SYSTEM’ ...................... 873
Choon Kiar Lim and S k n d r a Ranganarh, National University of Singapore, Singapom
POSTURE RECOGNITION OF NUCLEAR POWER PLANT OPERATORS BY SUPERVISED LEARNING. ................. .877
Cliikahito Nakajitna, Central Research Insrirure of Elearic Power Industry, Japan
ACTION MODELING WITH VOLUMETRICDATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 881
Fahio Cuuolin, Augur0 Sarri and Srefano Tubam. DEI - Polirecnico di Milano, Italy
MODEL BASED ALGORITHM FOR SINGULAR POINT DETECTION FROM FINGERPRINT IMAGES
Nmnan Wu and Jie Zhou. Deparhllrnr of Automation,Tsinghua University, China
ON-LINE WRITER RECOGNITION FOR THAI BASED ON VELOCITY OF BARYCENTER OF PEN-
Pifak Thumwarin and Takenohu Marsuura, Tokai University, Japan
M-BAND WAVELETS APPLICATION TO TEXTURE ANALYSIS FOR PALMPRINT RECOGNITION ..................... ,893
Dai Qingyun, Bi Ning, Huang Daren, Zhongshan University, China; David Z/lmg, Hongkong Polyu, Hong Kong; and Li Feng,
z/iOngS/~InUniversity, China
xii
REDUCED-COMPLEXITY BIOMETRIC RECOGNITION USING I-D CROSS-SECTIONS OF CORRELATION FILTERS. . . . . , 8 9 7
Juson Tliornton and Vijowkurnar Bkagavatuln. Camegie Mellon Univer.7it.y. USA

SESSION MP-P4: Image Segmentation: By Multiple Features and Other Methods

ADAPTIVE SKIN DETECTION USING MULTIPLE CUES ..... ....... . . . . . . . . . . .901
JinJeng Yang, Zlroirvrt Fu, Tirniir l i i n unrl Weiming He, Insrimre of Aukmiation, Chinese Acudemy of Science, National I~~rborarory of
Pattern Recognition, China
EFFICIENT FUZZY-CONNECTEDNESSSEGMENTATION USING SYMMETRIC CONVOLUTION AND
ADAPTIVETHRESHOLDINC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sliu-Yen Wim, Jimg-Eli Clien and Slru-Hung Yeh, Clrang Gung Universiry, Tniwan
IMAGE SEGMENTATION VIA BRITI'LE FRACTURE MECHANICS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..YO')
Wei W m g cmd Ronald Chsng, DPpt. of ACAE, Chinese Unia ojHongKong, Chinii
DIMENSIONALITY REDUCTION IN HYPERSPECTRAL IMAGE CLASSIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 I 3
Huiwen Zeng and H. Joel 7rimrl1, ECE Dept, NC State Unit: USA
AUTOMATIC SEGMENTATION OF BRAIN MRI THROUGH LEARNING HY EXAMPLE.. ............................. .917
Horacio 1,qaI-Awla nnd Jarques Fmon, Pontificirr Univer.yidode Calolicrr do Parano - PPCIA, B r u d
PERCEPTUALLY-TUNED MULTISCALE COLOR-TEXTURE SEGMENTATION . . . . . . . . . . . . . . . .921
1'oppn.i. Norrliwrslur-n Ihivcvsir?, USA: Aleksmdr-ri gowirz. IBM 7:J.

MULTI-SCALE SPACE VEHICLE COMPONENT IDENTIFICATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Willirsn Wai I.cutig ken, Cloncnt Clurn Clieong Pung and Nehon Hon Cliitiy f i q , 7%e Universi~?of Hortg Kong.Hmq Kong
OBJECT RECOGNITION BASED ON BINARY PARTITION TREES . . . ......... . . . . . . . . . . Y2Y
Orrste Snlemo. Montse Pardn.5, Veronica Vilciplana and Ferrun Marques,
EFFICIENT PROPOSAL DISTRIBUTIONS FOR MCMC IMAGE SEGMENTATION . . . .
Emo Kosriainm,Helsinki 1Jniver.yitv <$Teclinolqy, Laboratory of Computational Engineering, Fiiilatid; " I Jouko ,!"linen.
Hrlrinki Univer.iie of Tec1inolog~:Loboraru~ of Computational Engineering, Finlund
CONTOUR TRACKING BY MINIMAL COST PATH APPROACH. APPLICATION TO CEPHALOMETRY . . . . . . . . . 937
Burbarn Romaniuk, GREYC CNRS UMR 6072, France; Michel Desvignes, LIS, France; Marinetre Revenu, GR S UMR
6072, France: and Marie-Jos'ephe Desliu)'es, T'el'eCrhne Innovation, France
TOWARDS UNSUPERVISED ON OBJECT EXTRACTION BY INTEGRATING VISUAL ATTENTION AND
OBJECT GROWING. . . . . . . . . .......................................................................... 941
Junwei Han, School oJCompurer ring, Nangyung Tectmological University, Singapore;'King N. Ngan, Deparrtnent of Electronic
Engineering, The Chinese Universiry of Hong Kong,Hong Kong; Mingling Li and Honp,jinng Zhang, Microsoft Research Asia, China
IMAGE SEGMENTATION BY COOPERATIVE OPTIMIZATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,945
XiaoJei Huang, Callvista, Inc., USA
A CONNECTIVITY SOLUTION FOR EXTRACTION OF THIN OB1 ......... . . . . . . . . . .Y4Y
Marc0 Antonelli, Silvana Dellepiane, Cinnni Vemaua and Lorena Nov Biophysica ctronic
Engineering, 1tal.v
3D-COLOR-STRUCTURE-CODE - SEGMENTATION BY USING A NEW NON-PLAINNESS ISLAND HIERARCHY
Patrick Stunn, University of Koblenz, Germany

SESSION MP-PS: Image Enhancement 11

ADARIVE WAVELET RESTORATION OF NOISY VIDEO SEQUENCES .......................... 957
Nasir Rajpoor, Zhen Yuo and Roland Wilson, Universiry of Wanvick, UK
DIGITAL IMAGE INPAINTING USING MONTE CARLO METHOD . . . . . . . . . ,961
Jinnping Cu, Silong Peng and Xuelin Wang, Institute of Automation.
ADAPTIVE SPATIO-TEMPORAL FILTERING FOR VIDEO DE-N .......... Y 65
Hye-Yeon Cheong, Columbia University, USA; Alexis Michael Tourapis, Joan Llacli and Jill Boyce, Tlzomson Corporate Research, USA
THEORETICAL ANALYSIS OF SOME REGULARIZED IMAGE DENOlSlNG METHODS .............................. ,969
Patrick L. Combenes and Valerie R. Wajs, Universit'e Paris 6, France
IMAGE FUSION BASED ON NON-NEGATIVE MATRIX FACTORIZATION . ..... ....... 973
Junying Zhang, Electronics Engineering Instirule, Xidion Universiry, China; Le Wei, Scliool of Computer Science, Xidian University,
China; Qiguang Miao, Cuilin Institute of Electronic Technology, China: and Yue Wang, Department of Electrical and Compurer
Engineering, Virginia Polytechnic Institute and State Universio, USA
IMAGE DENOISING USING FREBAS MULTI-RESOLUTION IMAGE ANALYSIS ..................................... ,977
Saroshi Ito, Craduaie School of Eng. Utsunomiya Univ., Japan: and Yoshifumi Yamada, Faculty oJEng. Ursunomiyo Unix Japan
CONCEALMENT OF INTERPOLATION ERRORS FOR LOW BIT-RATE MOTION-COMPENSATED INTERPOLATION 981
Yon Wu, M.N.S.Swu,ny and M. Omair Ahmud, Concodia University. Canada

...
XI11
LEAST-SQUARES INTERBAND DENOlSlNG OF COLOR AND MULTISPECTRAL IMAGES. ........................... 98.5
Paul Sclieunders and JrfDriesen, Universifi <$Antwerp. Belgiarn
CONTRAST ENHANCEMENTOF RADIOGRAPH IMAGES BASED ON LOCAL HETEROGENEITY MEASURES.. . . . . . . . . ,989
Idris El-Feghi, Maller Sid-Ahtned, Mijid Ah,,tadi and Songrao Huang. University of Windsoc Canada
VISUALCONTENT ADAPTATION FOR LOW VISION USERS IN MPEG-21 FRAMEWORK
Truong Cong Tliung and Yong Man Ro. Information and Cmrmunicarion Univemity, South Korew
UNIVERSAL MINIMAX BINARY IMAGE DENOlSlNG UNDER CHANNEL UNCERTAINTY
and Tsarhy Weiwnan, Srnnford Univcrrifi. USA
George Ge,nelos, S ~ n r i i .Sigl~rjons.~on
r
A FAST AND ADAPTIVE METHOD FOR IMAGE CONTRAST ENHANCEMENT.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IN1
&yun Yu and Clrandrajit Bujuj, Departrrtmt of Computer Science, Uniwrsifi ofTe.ras ut Ausrin, 1J.U
SHIFTINVARIANT WAVELET DENOlSlNG USING INTERSCALE DEPENDENCY. .................................. .~100.5
Pei Clien rind David S~rlecDepl ECSE, Monash University, Ausrrdia

SESSION MP-P6: Video Object Tracking

MULTIPLE VIEWTRACKING OF HUMANS MODELLED BY KINEMATIC CHAINS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IOOY
Aravind Smdaresrm. Uniwni& of M u ~ ~ l r mCollege
d Purk, USA; Amir R o ~ ~ C I ~ o w r lUnioerrit?
lnq q"Cdiforniri, Riverside. USA:
ond R m m Clzclluppu, Oniorrsity $Maryland, Collcgr Pork, USA
MULTI-RESOLUTION PARAMETRIC REGION TRACKING FOR 2D OBJECT REPLACEMENT IN VIDEO. . . . . . . . . . . . . . . . 1013
Pml Hrasnerr. Duvirl Bull und N i 3 m Canagurajd~,Universifi of Bristol, UK
ROBUST SHAPE BASED TWO HAND TRACKER.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1017
Keran Horhore, Kririkul Solutions P V I . l.rd., Indiu; Kuustsblr Putwordlmn, S~tmanrmDirtin Rqy, Sublmsis Clmrcdhwi, Indim lnsiiiurr
of Technolog?: Bombay, India: und Sanranu Chaudliu: Indian lnsrirure of Technology. Dellzi. India
FACE CONTOUR TRACKING IN VIDEO USING ACTIVE CONTOUR MODEL.. . 1021
Xiong Ring, Yu Wei uninrl Clraruensuk Charawphun, EEE,NTU, Singupore
AIR-BORNE APPROACHING TARGET DETECTION AND TRACKING IN INFRARED IMAGE SEQUENCE . . . . . . . . . . . . . . 1025
Mukesh 7*riari, Shabbir Merchunr and Udav Desai, lndiuri Insrirnte of T ~ ~ l , ~ ~ l o g ? - B ~ , , ,hidiia
ba?,
A FAST AND ROBUST SIMULTANEOUS POSE TRACKING AND STRUCTURE RECOVERY ALGORITHM
FOR AUGMENTED REALITY APPLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1029
Ying-Kin Yu, Kin-Hong Wong, Department of Coinpurer Science and Engineering, Tlie Clr
and Micliael Ming-Yuen Chang, Department of Infunnation Engineering. The Clrinese University of Hang Kong, Hang Kong
A MULTICAMERA FUSION FRAMEWORK FOR MULTIPLEOCCLUDINGOBJECTS TRACKING I N
INTELLIGENT MONITORING AND SPORT VIEWING APPLICATIONS ............................................. 1033
Luca Marchesolti, Gianni Vemazza and Carlo Regazzoni. DlBE - Universifi of Genoa, Italy
BAYESIAN INTEGRATION OF A DISCRETE CHOICE PEDESTRIAN BEHAVIORAL MODEL AND IMAGE
CORRELATION TECHNIQUES FOR AUTOMATIC MULTI OBJECTTRACKING 1037
Santiago Venegas - Martinez, Gianluca Anronini, Jean-Philippe Thiran, Signal Processrag nstrrurr
Michel Bierlaire, Operation Research EPFL, Switzerland
ON-LINE PREDICTIVE APPEARANCE BASED TRACKING ...................................................... 1041
Namita Cupta, Poojo Mirtal, Indian lnsrirure of Technology, Delhi, India; Kausrubh Patwardhan, Sumanmm Durra Roy, Indian lnsrirure
of Technology, Bombay, India; Santanu ChaudhuT and Subhasis Banerjee. Indian lnsrirure of Technology, Delhi. India
A PROBABILISTIC COOPERATION BETWEEN TRACKERS OF COUPLED OBJECTS 1045
/do Leichtez Michael Lindenbaum and Ehud Rivlin, Technion - Israel Institute of Technology Israel
A TRAJECTORY-BASED BALL DFTE(JTION AND TRACKING ALGORITHM IN BROADCASTTENNIS VIDEO .......... 1049
Xinguo Yu, Institute for lnfocomm Research, Singopore; Chem-Horng Sim, National University of Singapore, Singapore; Jenny Ran
Wang. School of Compurer Science and Engineering, The University of New South Wales, Sydney 2052, Australia; and Loong Fah
Chcong, National University of Singapore, Singapore
SIMULTANEOUS BACKGROUND AND FOREGROUND MODELING FOR TRACKING IN SURVEILLANCE VIDEO 1053
Jie Shao, Shaohua Zlrou and Ram0 Chellappo, University ofMaryland, College Park, USA
MULTI-CAMERA CORRESPONDENCE BASED ON PRINCIPAL AXIS OF HUMAN BODY. ............................ 1057
Min Hu, lnsirirure ofAuromarion, ChineseAcademy ofSciences, China; Jiangwrng Lou, Microsof7 Research Asia, China: Weiming Hu
and 7ieniu Tan, lnsititure ofAuromarion, Chinese Academy of Sciences, China

SESSION MP-W: Biomedical Image Processing: Compression and Registration

APPLYING BINARY PARTITIONING TO WEIGHTED FINITE AUTOMATA FOR IMAGE COMPRESSION 1061
Kai Yang and Ghim Hwee Ong, School of Compuring, Norioml University of Singapore, Singapore
PROCESSING OF WAVELET TRANSFORM DATA FOR IMPROVED IMAGE COMPRESSION .......................... 106.5
Tanieem Muzaffar and Toe-Sun Choi, K-JISZ Korea
A NOVEL BOUNDARY HANDLING SCHEME FOR ARBITRARY OBJECT SHAPE IN VIDEO AND IMAGE COMPRESSION
Ka Shun Carson Pun and Truong Q. Nguyen, Depurtmenr of Elecrrical and Compurering Engineering, University of California,
San Diego. USA

xiv
LEAST-SQUARES MESH MODEL FOR IMAGE COMPRESSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1073
Iciar AIvnre;-Cnscos, Escriela Ticnicu Superior de Ingenieros de Trleco,,lunicocib,,, S p i n ; ond Yongyi Yrng. Illinois lnsritarr of
reclmologv, USA
RECONSTKUCTION OF 3U TOOTH IMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1077
Stephanie Buchaillard. Sin-Heng Ong. Notional Universiry of Singapore, Singupore: Yohon
France; and Kelvin Foong, Narional Universir). of Singapore, Singapore
SOFT SHAPE CONTEXT FOR ITERATIVE CLOSEST POINT REGISTRATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1081
Daeirl liu and Tsuhon Clien, Camegie Mellon Universin: USA
SHEAR-RESIZE FACTORIZATIONS FOR FAST MULTI-MODAL VOLUME REGISTRATION ........................... I085
Ying Clien, I'engwei Hao, Queen Man: Universir? of London. UK: and J i m Yu, Beijing Jiaorong Universily Chino
TECHNIQUES FOR TEMPORAL REGISTRATION OF RETINAL IMAGES . . . . . . . . . . . . . . . . . . . . . 1089
Bin Fang. Department <$Cornpurer S<;iencr,HKBU, Hong Kong; wme Hsir a n d Mongli Lee, Depn
NUS, Sinyopore

SESSION MP-PB: Video Coding

RATE-DISTORTION MODELING OF SCALABLE VIDEO CODERS.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I093
Min Dai. Dmitri Loguinoi! Texis A & M IJniwr.siry, USA; and Hayder Rodlra, Miclrigon Slate Unioer.sih, UXA
ON DECODEII-LATENCY VERSUS PER I N C . . . . . . . . . . . 1097
Pirilask 1,slmar and Konnun Ramr1x"an
LOWPASS FILTERING OF RATE-DISTO
VIDEO RECOKIIING AND STREAMING . . . . . . . . . . . . . . . . . ................................ I101
Utihoi He, Universiry of Missouri. Colunhia, USA; Chongwen Clren. Floridrr Insr. of7ixI1, USA; and Jianfei Cai, Nanyang
Trchnvlvgical U,,, Singapore
OBJECT-BASED VIDEO CODING USING A DYNAMIC CODING APPROACH.. ...................... I105
Marc C/ia~i,nont,IRISNINRIA, Frrtnce; St 'eplrone P o t c ~ ~Fmnce
r, Tiidcoin, France; and Henri Nicolas, IRISAN
MODEL-BASED MPEG COMPRESSION OF SYNTHETIC VIDEO SEQUENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1109
Davide Quaglia and Angel0 Garruso, Polirrcnicu di Torinn, Iralj
DIRECI N-POINT DCT COMPUTATION FROM THREE ADJACENT N/3-POINT DCT COEFFICIENTS . . . . . . . . . . . . . .
Soo-Cliang Pei and Meng-Ping Kao, Narional Taiwan Universily, Taiwan
REGION-BASED CODING OF MOTION FIELDS FOR LOW-BITRATE VIDEO COMPRESSION ......................... I I17
Huipin Zhang and Frank Bossmn, DuCoMo Communicorion Lobs USA Inc., USA
PROGRESSIVE POLYGON ENCODING OF SEGMENTATION MAPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1121
Marc Semais. Theo Vlachos. Universily of Surrey, UK; and Tlromas Davi
A JOINT DESIGN OF DICTIONARY APPROXIMATION AND MAXIMUM ATOM EXTRACIION FOR FAST
MATCHINGPURSUIT ....................................................... . 1125
Jian-Liang Lin, Wen-Liang Hwang, Institute of Infunnarion Science. Academia Sinica, Taiwun, R.O.C., Taiwan: and Soo-Chang Pei,
Depnrrment of Elecrricol Engineering, Narional Taiwan University, Taiwan, R.O.C.. Taiwan
IMPROVED DICTIONARIES FOR GENERALIZED BITPLANES-BASED MATCHING PURSUITS VIDEO
CODING USING RIDGELETS ................................................................. 1129
Leonard0 Fononteles, Pmgrama de Engenharia ElhricdCOPPE - Universidade Federal do Rio dr Janeiro, Brazil; Rogirio Caerano,
Fundacao Desembargador Pnulo Feiroza, Brazil: and Eduardo Da Silva, Pmgrama de Engenharia ElPtriCn/COPPE - Universidade
Federal do Rio de Janeira, Brazil
CONSTANT QUALITY RATE-CONTROLFOR VIDEO ENCODING BASED ON ACTIVITY SEGMENTATION , . . 1133
Luk Overmeire, Vlaamse Radio en Televisie (VRT). Belgium; Fabio Verdicchio, Joeri Burburien. Perer Schelkens, Vrije
Brusel, Belgium: and Lode Nachrergaele, Inreruniversity MicmElecrmnics Cenrer Belgium
RATE-DISTORTION OPTIMAL JOINT MACROBLOCK MODE SELECTION AND MOTION ESTIMATION
FOR MPEG-LIKE VIDEO C .................................................... I137
Hyungioon Kim and Yucel AI gy. USA
EFFICIENT CODING MODE DECISIONIN MPEG-4 PART-IO AVUH.264 MAIN PROFILE . . . . . 1141
lnchoon Choi, Jeyun Lee and Byeungwoo Jeon, Sungbunkwan University, Korea
SUBJECTIVE ASSESSMENT OF H.ZWAVC VIDEO FOR LOW-BITRATE MULTIMEDIA MESSAGING SERVICES.. . . . .
Patrick Brun, Cerl Hauske and Thomas Stockhammez Munich University of Technology, Germany

SESSION TA-S1: Content-basedAnalysis of Multi-modal High Dimensional Medical Images

A 3D MODEL OF THE HUMAN LUNG WITH LUNG REGIONS CHARACTERIZAITON 1149
Taljana Zrimec, Sara Rusayarar, University of New South Wales, Aurmlia; and Perer Wilson,
A STATE SPACE APPROACH TO NOISE REDUCTION OF 3D FLUORESCENT MICROS .......... 1153
Rairnund O6er; Xuming Lni, Universiry of E x a s ai Dallas, USA: Zliiping Lin, Nanyang Technological University, Singapore: and
Sally Ward, University of Texas Sourhwesrem Medical Center; USA

xv
AUTOMATIC EXTRACTION OF SEMANTIC CONCEPTS IN MEDICAL IMAGES I 157
Mira Park and Korngiri Rnitirr,eolranoroo, Tlte Uniwrsih, of Melbourne, Australin

SESSION T A 4 2 Image Forensics

ON INFORMATION HIDING WITH INCOMPLETE INFORMATION ABOUT STEGANALYSIS .......... . . . . . . 1161
Rajararlmcmt C l t m ~ l r m m d iSrevens
, Instilute of Teclmolagy, USA
STEGANALYSIS OF QUANTIZATION INDEX MODULATION DATA HIDING ...........................
Kennerh Sulliwin, Zltiqiung Bi, Upn,nanyu Mcidltow: Sliivkrsnar Cliandmsrkaran arid B.S. Manjunarh, Depr. of ECE,
Sanra Barbara, USA
A MODEL FOR IMAGE SPLICING.. . . . . . . . . . . . . . . . . . . . ...................................... 1169
Ean-Tsong Ng and Slrili-Fu Cltang. Dep'orrteenr of Elecrricd Engineering. Columbia Urtiversir?, New York, USA
A STOCHASTIC QIM ALGORITHM FOR ROBUST. UNDETECTABLE IMAGE WATERMARKING. . . . . . . . . .
Pierre Moulin and Alcxia Briassoudi,Universih, of Illinois, USA

SESSION TA-L1: Feature-based Image Segmentation

CORRIDOR SCISSORS: A SEMI-AUTOMATIC SEGMENTATION TOOL EMPLOYING MINIMUM-COST
CIRCULAR PATHS . . . . . . . . . . . . . ......................................................................... 1177
D k k FaJ-in, EdiiiiscIv Univcr.yireifE oven. The Nerlrerlands: Mr,gnirs Pfelfe,: Universiry Mn,tnlreiiri. German?: Pcrer De Wirlr,
Ecknirdte Universireir Eindlioven, Tlw Nerlrrrlnnds: and Wofgang Effdyberg. Uniw-.rih,Maiiiilwbti. Geniwn?
SKCS - NEW KERNELFAMILY WITH COMPACT SUPPORT.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1181
Ezzedine Ben Brairk, &%SIT Emisin; and Moltaiiied Clrcrier, ETS, Caiiadu
AUTOMATIC GENERATION OF PEN-AND-INK DRAWINGS FROM PHOTOS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 I85
Jialao Song. Tlie Hotig Kong Polyrecltnic Universiry, Hong Kong: Zhejiang Unharsirs, Hungzhou 310027, P R. China, Hong Kony;
Z l w u Chi, The Hong Kong Polyrechnic Universin: Hang Kmg, Hong Kong: Jilin Liu, Zlrrjiang l/niver.sih, Hangzlmrr 310027, P R.
China, Cltina: and Hong Fu, Tlie Hung Kung Polyrechnic Universin, Hong Kang, Hong Kong
A UNIFIED ADAPTIVE APPROACH TO ACCURATE SKIN DETECTION . . .
Qiang Ziu, ECE, UCSB. USA; Clring-Tung Wu, CS, UCSB, USA; and Kwang-Ting (Tiin) Clieng, ECE, UCSB, USA
AN INFORMATION THEORETIC FRAMEWORK FOR IMAGE SEGMENTATION.. . . . . .
Jaume Rigau. Miquel Feiras and Mareu Sberf, lnstirur d'lnfonri'alica i Aplicncions, Spain
TEXTURE ANALYSIS USING ADAFTIVE BIORTHOGONAL WAVELET P A C m S . . . . .
C. C. K . Ablwyararne, Electronic Engineering, Queen MaT, Universir? of London, UK; Ian H.
Sophia Antipolis. France
MODULATION-FEATURE BASED TEXTURED IMAGE SEGMENTATION USING CURVE EVOLUTION . . . . . . . . . . . . . . . . . 1201
lasonas Kokkinos, Georgios Evangrlopoulos and Perros Maragos, Narional Technical Universiry of Alltens, Crheece
DECOMPOSITION OF RANGE IMAGES USING MARKOV RANDOM FIELDS. ...................................... 1205
Andmas Pichler: Profactor Produkrionsforschung Gmbh., Austria: Roben B. Fishes School of Informarics. University of Edinburgh
UK; and Markus Vincze, Auromation and Conrrol Insrirure, Ausrria

SESSION TA-L2: Denoising and Deblurring

NOISE REDUCTION OF IMAGES WITH MULTIPLE SUBBAND TRANSFORMS. ..................................... 1209
Toshiliisa Tanaka, Tokyo University of Agriculrure and Technology, Japan; and Laurenr Duval, IFF: France
GEOMETRICAL IMAGE DENOISING USING QUADTREE SEGMENTATION . . . . . . . ............ ......... 1213
Rahul Shukla and Marrin Vetterli. Swiss Federal Institute of Tdinology Lausaune (EPFL).
FULL BLIND DENOlSlNG THROUGH NOISE COVARIANCE ESTIMATION USING GAUSSIAN SCALE
MIXTURES IN THE WAVELET DOMAIN ........... ......................................................... 1217
Javier Portilia, Universidad de Granada, Spain
MOTION-COMPENSATED TEMPRAL PRE-FILTERING FOR NOISE REDUCTION IN A VIDEO ENCODER .... 1221
Byung Cheol Song and Kong Wook Chun. Digital Media R&D Cenrer: Samsung Electronics Co.. Ltd, Republic of Korea
ASTROPHYSICAL IMAGE DENOISING USING BIVARIATE ISOTROPIC CAUCHY DISTRIBUTIONS IN
THEUNDECIMATEDWAVELETDOMAIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I225
A h Achim. Diego Herranz and Ercan Kuruoglu, Istiruro di Scienza e Tecnologie dell 'Infonnnzione, Italy
A NOISY CHAOTIC NEURAL NETWORK APPROACH TO IMAGE DENOlSlNG ........... ......... 1229
Leipo Yan, Lip0 Wang and Kim-Hui Yap, Nanyang Technological University, Singapore
FAST RELATIVE NEWTON ALGORITHM FOR BLIND DECONVOLUTION OF IMAGES . . . . . 1233
Alexander Bronsrein, Michael Bmnstein, Michnel Zibulevsky and Yehoshwr &mi, Technion - Israel lnsrilure of Technology, Israel
SHARPENING-DEMOSAICKING METHOD FOR REMOVAL OF IMAGE BLURS CAUSED BY AN
OkTICAL LOW-PASS FTLTER ................................................
Takashi Komarsu and Takahim Saito, Knnagawa University, Japan

xvi
 SESSION TA-W: Biometrics I1
FINGERPRINT IMAGE QUALITY ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1241
Eyung Lbn, Centre for Signal Processing, Singapore; Kar-Ann Toh, Insrimre for InJocom~nRe.search, Singapore; P.N. Sugnnrhnn,
Nunvang Teclznological Universir?, Singapore; Xtrdong J i m g and Wei- Yun Yau, lnsrirure for lnfo~o~nnz Research, Singapore
COMBINING EXCLUSIVE AND CONTINUOUS FINGERPRINT CLASSIFICATION ........................ 1245
Lgrrig Slra and Xinoou Tang, The Cliinrse Universi9 of Hong Kong, Hong Kong
CHOOSING BEST BASIS IN WAVELET PACKETS FOR FINGERPRINT MATCHING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1249
Ke H u n g ond Selin Ai,iyenre, Miclrigan .%ore Univer.yiry, USA
FINGERPRINT IMAGE QUALITY ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1253
7iri Pung Chen, Xudong J i m g and Weiyicn You, lnsrirurr for Infocomm Research. Singupore
RECONSTRUCTION-FREE MATCHING FOR FINGERPRINT SWEEP SENSORS.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1257
Perrr Morgirer, Clit-isti'm Nurr, Hrnning Lorclz. Injineon Techriobgie.s AG, Gennon?; Frmk Wollhoff and Gerlrord Rigoll, Murricl8
Univrrri8 of Technolog\: Gmna,t?
CASCADING STATISTICAL AND STRUCTURALCLASSIFIERS FOR IRIS RECOGNITION . . . . . . 1261
ZlrPnrrn Son. Yunlrong Wmg, Tienirr Em and Jiuli Cui, Cenrerjw Biomerrirs Aarltenricarion ond Tesring,
Palrent Recopition. Insrirsre of Auromnriun. Clrincse Amrlentv of Sc&m
RECOGNITION OF HUMAN AND ANIMAL MOVEMENT USING INFRARED VIDEO STREAMS. ...................... 1265
' @,I Jiong ond Cinrl? Daniell, HRI. I~borrrrories, LLC. USA
A NEW FACIAL EXPRESSION RECOGNITION TECHNIQUE USING 2-D DCTAND K-MEANS ALGORITHM I269
hying Ma, Concordiu IJniverrirj: Crrimdo: Yegui Xiuo, Hirashinio Prefrcriiml Wornm :r 1Jniver.sin: Jupan; K. K/iorasorii, Concordia
arid Ruhab Ward, Uniwmity of Briri.slr Columbia, Cunnrlu
1Jniver.sifx C~tiiorl<r;

SESSION TA-L4: Lossy Image Coding

CURVED WAVELETTRANSFORM AND OVERLAPPED EXTENSION FOR IMAGE CODING . . 1273
Dernin Wung, Liong Zhong and Andre Vincent, Cornrnunicarions Research Centre, Canada
UNDER-SAMPLED AND OVER-SAMPLED PRE-/POST-FILTERS FOR BLOCK DCT CODERS .......................... 1277
Clfcnpjie Tu, Windows Digital Media Division, Micmsafr Corporadon, USA: Trac D. Trnn, ECE Departmm, The J d m s Hopiins
Univer,~ity,USA; and Jie Liang,School of Engineering Science, Sirnon Fruser University. Canada
SPARSE REPRESENTATION OF IMAGES WITH HYBRID LINEAR MODEL ......................................... 1281
Kun Huang. Allen Yang and Yi Ma. University ofIllinois. USA
THE SEAMLESSLY MULTIPLEXED EMBEDDED CODEC (SMEC) AND ITS APPLICATION IN IMAGE CODING . . . . . . . . . 1285
Jin Li, Microsoft. USA
IMAGE SET COMPRESSION THROUGH MINIMAL-COST PREDICTION STRUCTURES .............................. 1289
Ckia-Ping Clten. Chu-Song Chen, Academia Sinica, Taiwan; Kuo-Liang Chung, National Taiwan University of Science and
Teclmologv, Toiwan; Hsueh-I Lu. Academia Sinica, Taiwan; and Gregory I! Tang, Nationul Triwan Universiry, Taiwan
JPEG2000-BASED SHAPE ADAPTIVE ALGORITHM FOR THE EFFICIENT CODING OF MULTIPLE
REGIONS-OF-INTEREST ............................. . . . . . . . . . . 1293
Mahesh Subedac Lino Kararn, Arizona State Universily, USA; and Glen Abowleman, General Dynamics C4 Systems, USA
A HIGHLY SCALABLE SPECK IMAGE CODER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1297
Cui Xie, Japan Advanced lnstiture ofScience and Te
SHAPE-ADAPTIVE CODING USING BINARY SET SPLI'ITING WITH K-D TREES. ................................... 1301
James Fowlez Mbsissippi Srafe Universiw, USA

SESSION TA-L5: Wavelet Video Coding and Scalability I

WEIGHTED AVERAGE SPATIO-TEMPORAL UPDATE OPERATOR FOR SUBBAND VIDEO CODING ............ 1305
Christophe Tillier, Beatrice Pesquet-Popescu, ENST, France; and Mihaela Van dcr Schaar, Univ. California Davis, USA
ACCURACY-SCALABLE MOTION CODING FOR EFFICIENT SCALABLE VIDEO COMPRESSION . . . . 1309
Guillaume Boisson, Edounrd Francois, Thomson R&D, France: and Christine Guillemot, IRISA, France
BIDIMENSIONAL DICTIONARY AND CODING SCHEME FOR A VERY LOW BITRATE MATCHING
PURSUIT VIDEO CODER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1313
Fulvio Moschetri, Kazuo Sug Minoru Etoh, NTTDoCoMo. USA
AN INVESTIGATION OF 3D DUAL-TREE WAVELET TRANSFORM FOR VIDEO CODING ............... 1317
Beibei Wang,Yao Wang. Ivan Selesnick, Polytechnic University, USA: and Anrhony Verm, Milsubishi Elecrric Research Lab, USA
SCALABLE MOTION VECTOR CODING.. . . . . . . . . . . . . . . . . . . ............................. 1321
Joeri Barbarien. Adrian Munteanu, Fnbio Verdrcchio, Yiannis Andreopoulos, Jan Comelis and Perer Schelkens, Vrije Universireit
Brmsel, Belgium
SPATIAL SCALABILITY AND COMPRESSION EFFICIENCY WITHIN A FLEXIBLE MOTION COMPENSATED 3D-DWT. .. 1325
Nagira Mehrsereslir and David Taubman, The Universiry ofNew South Wales, Australia

xvii
AN EFFICIENTCONTENT-ADAPTIVE MC 3D-DWT WITH ENHANCED SPATIAL AND TEMPORAL SCALABlLlT . . . . . . . 1329
Nagira Mehr.sereshr and David '/itubman, The University of New South Wales, Ausrralia
SCALABLE CODING OF VARIABLE-SIZE MOTION VECTOR BLOCKS ...................... 1333
Davide Maesrroni, Augrisro Sarri, Mano Tagliasarcchi and Srefrrno Tubaro, D

SESSION TA-PI: Sterroscopic a n d 3-D Processing I

FAST MODEL BASED STEREO MATCHING USING SOAR.. . . . . ...................... 1337
Yusr$Ozrurk and Amind Sridhamn. San Diego Srare Univer.sir>', USA
A SIMILARITY-BASEDADAPTIVE NEIGHBORHOOD METHOD FOR CORRELATION-BASED STEREO MATCHING.. . . . 1341
Madain Perez, Pierre Bonner, Olivier Color and Francois Cabestning, USTL, France
ADAPTIVE STEREO MATCHING ALGORITHM BASED ON EDGE IETECTION ............................ 1345
Kun Wang, Tlre lnsrirure of Arr$icial Intelligence rind Roborics, Cltino
CAMERA CALIBRATION WITHOUT METRIC INFORMATION USING ID OBJECTS.. . . . . . . . . . . . . . . . . . . . 1349
Xiaochim Sno and Hassan Fmuoslr, Universin, of Senrml Florida. USA
STEREO-BASED HUMAN HEAD DETECTION FROM CROWD SCENES.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1353
Xiaoytc Hrtang, Liyuan Li, lnsrirurr for Infocomm Reseorclr, Singapore; and Terence Sim, Narional Univemin, ofSingr,pre. Singapore
UNCALIBRATED STEREO RECTIFICATION FOR AUTOMATIC 3D SURVEILLANCE.. . . . . . . . . . . . . . . . . . . 1357
Se,--Nanr Liiii, Lan? D m i s Univ of Ma~luiid,Cdlegc Park, SS dqJ1.. USA; Anrrrag Miool, Siemeris Corporare Reseorck, Pri,rcurmr,
USA: and Nikm Pamgim, Ecole Narionole de P0nr.s et Clraussea, Fmnre
ROBUST BAYESIAN CAMERA MOTION ESTIMATION USING RANDOM SAMPLING. ......................... 1361
Gang @an, Arizona Srnre Uniwrsil?, USA; Rrsna Chdlappa and Qinfen Zheng, Uniwrsiiy of Maryland, USA
SHAPE RECONSTRUCTION FOR COLOR OBJECTS USING SEGMENTATION AND PHOTOMETRIC S T E R E O . . . . . . . . . . . 1365
Osaeiu lkcda, Takusliokri Universiry, Japan
A ROBUST METHOD FOR RECOVERING GEOMETRIC PROXY FROM MULTIPLE PANORAMIC IMAGES.. . I369
Ada S.K. Wan. Angus M. K. Sir), Rynson W H. Lnu and C. W Ngo. Cin, University of Hong Kong, Hong Kong
DISPARITY ESTIMATION USING COLOR COHERENCE AND STOCHASTIC DIFFUSION ........................ 1373
Sang Hwa Lye, Nain I k Clto, Seoul National University, Sourlz Korea; and Jong-il Park, Hanyang Universit>j,South Korea
EFFECTIVE INTERPOLATION FOR FREE VIEWPOINT IMAGES USING MULTI-LAYERED DYNAMIC
BACKGROUND BUFFERS ...................................................... 1377
Atsuslri Marsumura. Sei Nai d Sliuichi Marsrunoto, KDDl R&D Laboratories Inc., Japan
BUILDING BLOCKS FOR AUTONOMOUS NAVIGATION USING CONTOUR CORRESPONDENCES . . . . . . . . 13x1
M. Pawan Kumas S. CI Jawahar and R J. Narayanan, CV17: 1117: India

SESSION TA-P2:Face Detection, Recognition and Classification I1

NAIVE BAYES FACFJNONFACE CLASSIFIER: A STUDY OF PREPROCESSING AND FEATURE
EXTKACTIONTECHNIQUES .............................................................. I385
Son Lnm Phung, Abdesselom Bouzerdoum. Douglas Chai and Anthony Watson, Edith Cowan University, Ausr
REFLECTANCE CORRECTION FOR PERSPIRING FACES ................................. 1389
William Smith. Antonio Robles-Kelly and Edwin Hancock, Universiry
A ROBUST FEATURE EXTRACTION FRAMEWORK FOR FACE RECOGNITION. ........................
Guang Dai and Yuniao Qian, College of Computer Science, Zhejiang University, China
SELF QUOTENT IMAGE FOR FACE RECOGNITION . . . . . . . . . . . . . . . . 1397
Hairao Wang. Insriftae of Automarion, Chinese Academy of Scien
Instirule of Automarion. Chinese Academy of Sciences, Chino; and Jian J. Zhang. NCCA, Media School, Boumemourh University, UK
FACE IDENTIFICATION FROM ONE SINGLE SAMPLE FACE IMAGE.. ....................... 1401
Hung-Son Le and Haibo Li, Depr. ofdpplied Physics and Electmnics, Umea University, Sweden
FACIAL EVENT MINING USING COUPLED HIDDEN MARKOV MODEL ......... 1405
Limin Ma, Qiang Dou, Mehmer Celenk and David Chelberg. Ohio Universiry, USA
FACIAL EXPRESSION ANALYSIS BY KERNEL EIGENSPACE METHOD BASED ON CLASS FEATURES
(KEMC) USING NON-LINEAR BASIS FOR SEPARATION OF EXPRESSION-CLASSES ................................ 1409
Yohei Kosaka and Kazunori Korani, School of Informarion Science, Japan Advanced Instirule of Science and Technology, Japan
FACE DETECTION TECHNIQUE BASED ON INTESITY AND SKIN COLOR DISTRIBUTION ....................... 1413
Satyanadh Cundimada, Li Tao and VijayanAsari, Old Dominion University, USA
FUSIONOFSVDANDLDAFORFACERECOGNITION .......................................................... 1417'
Yanwei Pang, Rong Zhang, Zliengkai Liu, Microsoft-USTC Intelligent Computing Research Cenrer: China; Nenghoi Yu. University of
Science and Technology of China, China; and Jiawei Rong, Fudan University, China
THREE-DIMENSIONAL FACE RECOGNITION: AN EICENSURFACE APPROACH 1421
Thornas Heselrine, Nick Pears and Jim Austin, The Uniwmin, of York, Britain

xviii
FAST FACIAL FEATURE EXTRACTION USING A DEFORMABLE SHAPE MODEL WITH HAAR-WAVELET
BASED LOCAL TEXTURE ATTRIBUTES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1425
Fei Zuo, Eitidlioven Universiy of Tedmology, The Nedter1and.y; on nul. / LogicaCMG. The
Netherlands
CURVATURE RASED HUMAN FACE RECOGNITION USING DEPTH WEIGHTED HAUSDORFF DISTANCE. . . . . . . . . . . . . 1429
Yeung-Hak h e , Yeitngnnm Universiy, Soirtli Korea; and Jae-Cliong Slriin. Andong Nrrtiunal Uniserri?, South Korea
FACIAL RECOGNITION AND VERIFICATION USING GABOR WAVELETS AND KERNEL METHODS . . . . . . . . . . . . . . . . . . 1433
Li Hoi, Nuttingknni Univar.sir\: UK; Linlin She,,, Univeniy ofNottingham, UK; and Phil Picrun, Universiy C o l l q e Nortlzompon, 1JK
PEOPLE MONITORING USING FACE RECOGNITION WITH OBSERVATION CONSTRAINTS. ......................... 1437
Ji RIOand Yap-Peng Em. N o n p n g Technologicrrl UniversirJ Singupore

SESSION TA-P3: Motion Detection and Estimation: Block Matching

A MULTISTAGE FAST MOTION ESTIMATION SCHEME FOR VIDEO COMPRESSION. ............................... 1441
Jiancony Luo, Isl@y Alsnad, Yu Sun mid Yondong Lions, The UniversiF of Texas at Arlington, USA
FAST BLOCK-MATCHING MOTION ESTIMATION BY RECENT-BIASED SEARCH FOR MULTIPLE REFERENCE FRAMES . . 1445
Chi- Wang 7 h g , Wing-Hung L a m and Lni-Mm Po, C i y IJnivrrriy ofHong Kong. Hong Kong
SHORT/LONG-TERM MOTION VECTOR PREDICTION IN MULTI-FRAME VIDEO CODING SYSTEM . . . . . . . . . . . . . . . . . . 1449
K-HOl7 H s b o , Tien-Hm h e and Pao-ChiClrang. Deporrmwt of E'lecrriml Eitgincering, Nritionol Cenrrol University,Trriwm
SEARCH WINDOWSIZE DECISION FOR MOTION ESTIMATION ALGORITHM IN H.264 VIDEO CODER . . . . . . . . . . . . . . . 1453
Gianluco Brdo. Mnssbnu Buriani, Ivuno Burbieri and Mrrrco Roygio, Dibe - University of Genoorr, /to/?
ADAPTIVE BLOCK SUB-SAMPLING ALGORITHM FOR MOTION-ESTIMATION ON SlMD PROCESSORS.. . . . . I457
Munish Jitidirl rind Nagr.yw,or[r Roo C., Emirzed India Pvr. Ltfl., India
COMglNED ?D OBJECT MOTION ESTIMATION I N MEDICAL SEQUENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1461
Delliiry Bertrand. Clarysse Potrick, CREATIS, France; Grungeor Pierre, CEA-LET nce: Magnin Isobellc, CREATIS. Frmce: md
Bonnet Stephaiie. CEA-UTI, Frrmce
RECURSIVE TEMPORAL DENOlSlNG AND MOTION ESTIMATION OF VIDEO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1465
Vladimir Zlokolica, Alrksundrn Pizarica and Wi!/"ied Philipx, TELIN, Univer.sir,y of Cheat, Belgiuvi
AN ADAPTIVE MOTION ESTIMATION ALGORITHM BASED ON EVOLUTION STRATEGIES WITH
CORRELATED MUTATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1469
Wang Hui and Mao Zliigung, Hrrrbin lnsriture of Technology Micmelecrronics Center 313, Cliino
NOVEL FRAME INTERPOLATION METHOD FOR HOLD-TYPE DISPLAYS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1473
Nu0 Mishinla and Goh Itoh, Multimedia Loboratoq, Corporare Reseorch & Develop Cenrec Toshiba Corp., Japan
A NOVEL FLAITED HEXAGON SEARCH PATTERN FOR FAST BLOCK MOTION ESTIMATION.. . . . . . . . . . . . . . . . 1477
Thou-Ho (Chao-Ho)Clzen and Yi-Fan Li, Department of Electronic Engineering, Nutional Kaohsiung University ofApplied Sciences,
Taiwan
ATWO STAGE VARIABLE BLOCK SIZE MOTION SEARCH ALGORITHM FOR H.264 ENCODER 1481
Tomoyuki Sliimizu, Akio Yuneyama, Himnosa Yarzagilraraand Yasuyuki Nahjimu, K D D I R&D Loboram
FAST FULL SEARCH BLOCK MOTION ESTIMATION FOR H.264JAVC WITH MULTILEVEL SUCCESSIVE
ELIMINATION ALGORITHM . . . . . . . . . . ... ............................. 1485
Tuukko Toivonen and Janne Heikkilii, University of Oulu.
FUZZY NON-RIGID MOTION ESTlMAnON ROBUST TO ROTATION .............................................. 1489
Juan Moroles Sanchez, Rafael Verdu Monedero, Ricardo Gonzalez Leon, Universidad Politecnico de Cartugena. Spain; and Luis
Weruaga Priero, Austrian Academy of Sciences. Austria

SESSION TA-P4: Feature Extraction and Analysis: Color and Texture

TEXTURE DISCRIMINATION USING MULTIMODAL WAVELET PACKET SUBBANDS 1493
Roberto Cossu. Ian Jemyn and Josiane Zerubia, INRIA. France
COMPARISON OF LINEAR SPECTRAL RECONSTRUCTION METHODS FOR MULTISPECTRAL COLOUR IMAGING . . . ,1497
David Connah, Jon Hardeberg, Cjpvik University College. Norway; and Steplten Westland, Leeds Univemi@, England
OBJECT TRACKING BY ADAPTIVE FEATURE EXTRACTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1501
Bohyung Hun and Lorry Davis, University ofMaqland-College Park. USA
COLOR TEXTURAL FEATURES UNDER VARYING ILLUMINATION. .............................................. 1505
Dimitrios Iakovidis. Univ. of Athens, Greece; Dimirrios Maroulis, Univ. oft Athens, Greece; and Stavms Karkonis, Technological Inst. of
Lamia, Greece
ADVANCES IN TEXTURE ANALYSIS: ENERGY DOMINANT COMPONENT & MULTIPLE HYPOTHESIS TESTING . . . . . . 1509
lasonas Kokkinos. Georgios Evangelopoulos and Petms Maragos. National Technicul Universiry ofAthens. Greece
A FAST PROCEDURE FOR THE COMPUTATION OF SIMILARITIES BETWEEN GAUSSIAN HMMS .................... 1513
Ling Chen and Hong Mun, Stevens Institute of Technology, USA

xix
MULTISCALE ASYMMETRY SIGNATURES FOR TEXTURE ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 17
Cerr Van de Wouwec Wsielab, UniversiQ of Anmerp. Relgiuin: Barbam Weyn and Dirk Van Dwk, Universiry of Anhwrp. Belgium
A NEW SVM KERNEL FOR TEXTURE CLASSIFICATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1521
Muhdi Subri und Juvud Alirezaie. Rverson Universify, Conudcr
A MEASURE FOR SPATIALDEPENDENCE I N NATURAL STOCHASTIC TEXTURES.. ..... ....................... 1525
Roherro Cosrunrini. Ecole Polyrechnique Federulr de Lnirsnnne, Swifzerlond: CIorirr Menrgrrz, Universiv of Sienu, Italy: and Srrhine
Sussrrunk, Ecole Polvtechnique Federrile de Lnusanne, Swirzerlrrnd
SEQUENTIAL UPDATING ALGORITHM FOR EXTRACTING THE BASIS OF THE KARHUNEN-LOEVE TRANSFORMATION. . 1529
Yanyun Qu, Depumnenr of Coinpurer Science of Xiuiiien Unherrify, China; Nanning Zlreng, Zejian Yuan, lnsrirure of Arfificiul
l n t e l l i p x x and Robotics OJXi in Jiaoriong Universiry. Cllino: and Cuilruu 1.i. Drparrmenr of Compurer Science of Xiranen Univerri@
Ckino
A NEW PALETTE HISTOGRAM SIMILARITY MEASURE FOR MPEG-7 DOMINANT COLOR DESCRIPTOR . . . . . . . . . . . . . IS33
Lni-Mun Po and Kct-Mun Wong, CiQ Unisersiry of Hong Kung, Hong Kong
TEXTURE SIMILARITY EVALUATION USING ORDINAL CO-OCCURRENCE .
Mori Purfio, Bogdon Crmiiuriuc and Moncef Gabbouj, Trrinperr Univer.siry of Techno
A COMPARISON O F THE OCTAVE-BAND DIRECTIONAL FILTER BANK AND GABOR FILTERS FOR
TEXTURE CLASSIFICATION . . ...............................
Puul Hong, Geoqia hrsrirrne ( f f i c rlmro Universif?, USA: om1 Mark Sinif
ROTATION INVARIANTTEXTURE CLASSIFICATION USING DIRECTIONAL FILTER BANK AND
SUPPORTVECTOR MACHINE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I545
Hung Man. Ling Chm and Rvng Dum. Srcoens lnsiirure of Technolog?: USA

SESSION TA-PS: Watermarking 111

PREDICTION-ERROR BASED REVERSIBLE WATERMARKING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I549
Diljirh Thodi und Jrffwy Rodriguez, The Universiry ofArizona, USA
AN IMAGE NORMALIZATION BASED WATERMARKING SCHEME ROBUSTM GENERAL AFFINE TRANSFORMATION . . 1553
Hwan 11 Kung. Myongji Universify, Sorrrli Koreo; and Edward Delp, Purdue Ofliversir):USA
DATA HIDING IN JPEG 2000 CODE STREAMS . . . . . ....... ....... ..... . . . . . . . . . 1557
We; Liu, Panusonic R&D Center China Co. Lfd.. Cliinu
SINR. BIT ERROR RATE, AND SHANNON CAPACITY OFTIMIZED SPREAD-SPECTRUM STEGANOCRAPHY.. . . . . . . . . . I561
Maria Ckizeli, Dirnifris Pados, Sfate Universiry of New York ur Buff&, USA; and Michuel Medley, SUNY Insrirure of Technology, USA
A SUITABLE IMAGE HIDING SCHEME FOR JPEG IMAGES ..... ..... ..... .... . . . . . 1565
Sliinfeng D. Lin, Shih-Chieh Slrie und Chung-Chien Chou, Depr. of Computer Science and Infortnorion Engineering. National Dong Hwa
Universiy, Taiwan
A BLIND ROBUSTWATERMARKING SCHEME FOR COPYRIGHT PROTECTION OF 3D MESH MODELS . . .

Jinhai Wu.Tsinghuu Univ., China: Bin Zhu. Shipcng Li. Microsoff Research Asia, China: and Fuzong Lin, Tsinghua Univ., China

Republic of Koreo; Young Huh. Korea Elecfrorechnology Reseorch Inrrirule, Republic of Korea; Ki-Ryong Kwon, Depunnienr of
Elecrmnic Engineering, Pusan Universify of Foreign Srudies, Republic (8Korea; and Kuhn-I1 Lee, School of Elecfricul Engineering &
Compurer Science, Kyungpook Norional Univenify, Republic of Korea
A NEW ALGORITHM FOR COLLUSION RESISTANT VIDEO WATERMARKING. .. ...... ....... . . 1581
Vinod P and P K Bora. Deparfmenr of ECE, IIT Guwahati, India

SESSION TA-P6 Video Indexing, Retrieval and Editing

GENERIC SLOW-MOTION REPLAY DETECTION IN SPORTS VIDEO ......... .... 1585
LeiWung, Xu Liu. Tsinghua Univemify, China; Steve Lin, Microsoff Research Asia, China; Guungyou Xu, Tsinghua Universify, China; and
Heung-Yeung Shum, Microsoff Research Asia, China
ABRUIT AND GRADUAL SCENE TRANSITION DETECTION IN MPEG-4 COMPRESSED VIDEO SEQUENCES
USINGTEXTUREAND MACROBLOCKlNFORMATlON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I589
U( Ani1 Chanduna Femundo and Kok-Keong Lao, Brunel University, U K
REAL-TIME TEMPORAL TEXTURE CHARACTERISATION USING BLOCK-BASED MOTION
CO-OCCURRENCE STATISTICS. . . . . . . ... ....... . . . . . . . . . . . . . 1593
Ashfaqur Rahmn and Muntur Murshed, Monush Uniwrsify, Awfraliu
MEAN SHIFl' BASED NONPARAMETRIC MOTION CHARACTERIZATION . . . . . . . . . . . . . . . . . . 1597
Ling-Yu Duan, Min Xu, Qi 7im and Chang-Sheng Xu, Insrirure fur lnfocomm Research, Singapore

xx
FAST VIDEO SHOT RETRIEVAL BY TRACE GEOMETRY MATCHING I N PRINCIPAL COMPONENT SPACE. . . . . . . . . . . . 1601
Zliu Li. Motomla LaDs. SchaurnDurg, USA; Aggelos Krrlsnggelos. Norrlrwsrern University, Evansron, USA: mid Bhuvan Gandlzi,
Molorola Irrbs. Sclramhurn. USA
AL"I'O\lI\fIC:\LLY I . t R U I \ C i S'fKUC'TL1K.IL U V l T i IU E I I ~ . C A I ' I ~ J Y , VI1)I'OS
4I. M'ITH THlr
IIIFIIAKCIIIC~ILHII)I)FN MARKOV I\lOI)LLS. . . . . . . . . . . . . . .
~ ~~
......... I605
Dinlr Pliung. S1.etlra Vmkaresli, Curtin University of Technology,Ausrmlirr: and Hung Bui, A I Cenre,: SRI lnreniorional, USA
ENHANCING LATENT SEMANTIC ANALYSIS VIDEO OBJECT RETRIEVAL WITH STRUCTURAL INFORMATION
Fubrice Souvannnvong, Lukrrs Hollz, Bemnrd Meriald(i orid Benoir Hue!, EURECOM, France
CONTENT AND TRANSFORMATION EFFECT MATCHING FOR AUTOMATED HOME VIDEO EDITING . . . . . . . . . . . . . . . . 1611
Xiun-Shcng Hut and Hong-Jiang Zlrnng. MicrosnJi Research Asin. Cllino
TIME-CONSTRAINT BOOST FOR TV COMMERCIALS DETECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1617
Fe- Em Liu, Micro.s,fI Researrh Asia. Chino: Tco Qin, Depr. Elrcrmnic Enginee,-ing, Tsinglum Univercir?, Clzinrr: and Hong-Jinng
Zliong, Micrnsnfl Resrordi Asia, Cliinri
SEMANTIC UNITS BASED EVENTS DETECTION IN SOCCER VIDEOS
Ximfenp Ring. Qingslrriii Liic and Hrmqing 1.u. N(irinnal InDorafop of Parre
Sciences, Chino
HIGH-LEVEL SOCCER INDEXING ON LOW-LEVEL FEATURE SPACE, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1625
liic, .Japm: Knichi Uemur-a, Scieizce Uniocr.siry (nf 7 b k w Jupun: Yast,wki Nakojimo
Martnr S ~ i g a n n .K D D l R&U Lnbo,omrori~,.s otrd
Hirotmw Yonogilrurrr. K D D l R& D Iriboru~orirsInc., J<ip,on
GOAL UETECTION IN SOCCER VIDEO USING AUDIO VISUAL KEYWOR ........................ 1629
Er-Lin K m g , Joo-Hwre I h , brsrirure for Infocumm Rrseo,rI!, ~ i l I g a p < J ' FMoho,
: 01 Universiry < ~ S i n g a p o , u ,
Singapore: Cliangsheng Xu rrnd Qi Tim, Insrirurcfor lnfocniiiin Research, Singapore
A SOCCER FIELD TRACKING METHOD WITH WIRE FRAME MODEL FROM TV IMAGES.. ......................... 1633
7irnoki Waronobe, Miki Hasewma and Hide0 Kirojiiim, Sclrool of Enginerring, Hokkoido Universiry, Jupun
EVENT DETECTION BASED ON NON-BROADCAST SPORTS VIDEO. . . . . . . . . . . . . . . . . . . ........................ 1617
Jinjun Wrmg, Clrongsheng Xu, lnsrirus for Infocoiim Resmrclr, Singapore: Engsiong Clzn, Nariyang Technologica/ Utriversily, Singupore;.
.Xinguo Yu om1 Qi Eon, 1n.srirrrrefor Infi~commReserrrdi, Singapom

SESSION TA-W: Interpolation

MMSE RECONSTRUCTION FOR 3D ULTRASOUND IMAGES.. ... ........................ 1641
Wei Huong, Ehin Zlieng and Janelle A. Molloy, University of Virginia. USA
QUANTITATIVE ANALYSIS OF RESOLUTION SYNTHESIS.. . . . . . ........................ 164s
Ramez Youkeim and David Toubman, University of New Sourh Wales, Au
NOISE REDUCTION AND INTERLACED-TO-PROGRESSIVE CONVERSIONBASED ON OPTIMAL ADAFTATION ....... 1649
Seungioon Yang. Joe-Han Jung and You-Young Jung, Samsung Elecrmnics Co., Lld, Korea
GENERIC METHOD FOR 2D IMAGE RESIZING WITH NON-SEPARABLE FILTERS
Christian Henrscliel, Brandenburg Universiry of Technology Conbus, Germany
A NORMALIZED MODEL FOR COLOR-RATIO BASED DEMOSAICKING SCHEMES.. . ........................ 16.57
Rusrirlav Lukac and Konslanrinos Plaranioris, Universiry of Tomnlo. Canada
LARGE-SCALE INFOGRAPHIC IMAGE DOWNSIZING. ........................
Ruihua Ma, Insrirure of Infococomm Research. (IZRJ, Singapore: and Gurminder Singh, Naval Posrgraduure School, USA
USING NATURAL IMAGE PROPERTIES AS DEMOSAICING HINTS.. .. ....... ................ 1665
Id0 Omer and Michael Wermo'n,HUJI, Israel
ROBUST CORRECTION STEP FOR CFA INTERPOLATION SCHEMES ................................. ... 1671
Rasrirlav Lukac, Karl Martin. Konsrantinos Plataniotis, University of Toronro, Canada: and Bogdan Smolkn, Silesiun University of
Technology, Poland
IMAGE MORPHING BASED ON MUTUAL INFORMATION AND OPTIMAL MASS TRANSPORT.. ..................... 1675
Lei Zhu, Yon Yang. Allen Tannenhaum, Georgia lnsrirure of Technology, USA: and Steven Hake,: Brighurn and Women>Hospital and
Haword Medical School, USA
A RECURSIVE APPROACH FOR DE-INTERLACING USING IMPROVED ELA AND MOTION COMPENSATION
BASED ON 61-DIRECTIONAL BMA ...... .......... .......... : . . I 6 7 9
Seungchan Byun, Jeongmoon Byun and Gyeonghwan Kim, Soga
MULTI-LEVEL FAST MULTIPOLE METHOD FOR THIN PLATE SPLINE EVALUATION . . . . . . . . . . . . . . . . . . 1683
Ali Znndifnc Ser-Nam Lim, Ramoni Duraiswami, Nail A. Gumemv and L n r p S.Davis. University of M a y l m d . USA
IMAGE INTERPOLATION BASED ON INTER-SCALE DEPENDENCY IN WAVELET DOMAIN ......................... 1687
Dong Hun Woo. Dept. Electronic Eng. Pusan National University, South Korea; II Kyu Eom, Depr. Information and Communicario Eng.
Miryang Narional University, Sourh Korea: and Yoo Shin Kim, Depr. Elecrmnic Eng. Pusan Narional University, Sourh Korea

xxi
SESSION TA-PS: Geosciences and Remote Sensing and Environment
MULTISCALE SEGMENTATION OF TEXTURED SONAR IMAGES USING COOCCURRENCE STATISTICS 1691
bnm Karoui, Jeun-Marc Bouclres ENST-Bretagne, Frrrnce; Ranan Fabler ond Jean-Marie Augustin, IFREMER, Fr
GENERATION OF VIDEO METADATA SUPPORTING VIDEO-CIS INTEGRATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169.5
In-Huk J m , Tar-Hym Hwong and KyunrHo Chui, Electronics and Telecom,nunicrrriuns Researcli Insrirute, South Koreu
ELEVATION CONTOURS EXTRACTION FROM A COLOR-CODED RELIEF SCANNED MAP . . . . . . . . . . . . . . . . . . . . . . . . . . 1699
Carol Rids. Cnrneliu Ruse, Trchnical Universitv of CLcj-Napncu, Roaiuniu; and Jaakko Astnlu. Tampere International Center /or Signul
Processing, Finlund
A REGION-BASED METHOD FOR GRAPH TO IMAGE REGISTRATION WITH AN APPLICATION TO CADASTRE DATA. . . . 1703
Roger Trios-Sanz and Marc P i ~ ~ ~ ~ ~ t - D ~ ~lnsritrrt
~ i l l i Grogmaplrique
g"y, National, France
AN EARLY FIRE-DETECTION METHOD BASED ON IMAGE PROCESSING . . . . . . . . . . . . . . . . . . 1707
Tlrov-Hu (Chuo-HoJ Clzen. Ping-Hsueh Wu und Yung-Chireii Clriuu, Deportnienr rtiunul Kuohsiung
University of Applied Sciences, Taiwcrn
A NOVEL TWOCSTEP STRATEGY FOR AUTOMATIC CIS-IMAGE REGISTRATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 I I
Zlronwu Yu, Notiunnl Lr,bnrarnn~'o/Pattern Recognition, 104CAS, Clzina; Veroniqice Priner and Clrunhang Pun. Natiorinl Lnbomrop of
Patiern Recongitiun,lOA. CAS. Cliina
NOVEL AIRCRAFTTYPE RECOGNITION WITH LEARNING CAPABILITIES IN SATELLITE IMAGES . . . . . . . . . . . . . . . . . . 1715
Jun Wei Hsiek, Dqxwtncenr of Electrical Engineering, Yuun 2 U,tiwr.ri?, TOiwu,~;Jiru-Ming Clwt. Chi-Hung C l i u a q and KrtoCHin
Fuu, Dqmrrm~ntof Co,ttpraer Engineering, Ncirionnl C o a d Univer.rity, 7iii,cnn
A LAND USE CLASSIFICATION METHOD BASED ON REGION AND EDGE INFORMATION FUSION . . . . . . . . . . . . . . . . . . 1719
Gang Wu, C l r ~ m h u qPun, Vwuniqrre Pririet and Sungdr. Ma, Nafionul Lr,bomro,y q'Prrtle,n Recrqytition,I,i,srirurr uJ'Auru,nation.
Cliiirese Academy n/Sciences, Cliino
SEMI-AUTOMATIC ROAD DETECTION FROM SATELLITE IMAGERY . . . . . . . . . . . . . . . 1723
S o d a i r Udomhansnkel. King Mongkut k Institute o/ Tcclrnology Lndkmbang, Tlinilanil
TOWARDS COMPUTER-ASSISTED PHOTO-IDENTIFICATION OF HUMPBACK WHALES ............................ 1727
Elena Ranguelnva. Mark Huiskes and Eric Pauwels, CWI, The Netherlands
BUILDING EDGE DETECTION BASED ON DYADIC WAVELET . . . . . . . . . . . . . . . . 1731
Qiiiiing Qin, lnstitute.ofRs&GIS, Peking University, China; Wenjun Wang. Deportment of Co,nmunicution Errginrering, Yanshnn
University, China; and Sijin Chen. lnsritute o/R.s&GlS, Peking University, Cliina

SESSION TP-S1: What is the Latest in Networked Video?

OI'TIMAL ERASURE PROTECTION FOR SCALABLY COMPRESSED VIDEO STREAMS WITH LIMITED
RETRANSMISSION ON CHANNELS WITH 11D AND BURSTY LOSS CHARACTERISTICS .................... 1735
Johnyon Thie and David Taubman. The Universiry of New South Wales, Australia
VIDEOCONFERENCING OVER AN INTERMEDIATE-PROXY ..................................................... 1739
Ali C. Begen and Yucel Altunbasak, Georgia Institute of Technology, USA
TRANSMISSION PROTOCOLS FOR STREAMING VIDEO OVER WIRELESS ........................................ 1743
Minghua Chen and Avideh Znkhor; EECS. U C Berkeley, USA
OVERLAY AND PEER-TO-PEER MULTICAST WITH NETWORK-EMBEDDED FEC .................................. 1747
Hayder Rudho and Mingqwrn Wu, Michigan State Universiry, USA
MINIMIZING DISTORTION FOR MULTI-PATH VIDEO STREAMING OVER AD HOC NETWORKS 1751
Eric Senon, Xiaoqing Zhu and B e d Girod, Stanford Universiry, USA
SCALABLE PREDICTIVE CODING BY NESTED QUANTIZATION WITH LAYERED SIDE INFORMATION .............. 1755
Huirheng Wung and Antonio Ortega, Signul und l m g e Processing Institute, Deportment of Electrical Engineering, University of Southern
California, USA
A PEER-TO-PEER VIDEO-ON-DEMAND SYSTEM USING MULTIPLE DESCRIPTION CODING AND SERVER DIVERSIT. . 1759
Xiaofeng Xu, Yao Wang. Sliivendra Panwar and Keith Ross. Polytechnic Universify, USA
SECURE MEDIA STREAMING & SECURE ADAPTATION FOR NON-SCALABLE VIDEO 1763
John Apostolopoulos, HP Lobs, USA

SESSION TP-L1: Super-resolntion and Interpolation

MOTION COMPENSATED SUPER-RESOLUTION OF VIDEO BY LEVEL SETS EVOLUTION. .......................... 1767
Curlos Vozquez, INRS-EM?: Canada; Hussein Aly, SITE, University of Ottawa, Canada; Eric Drbois, Faculty of Engineering. University
of Otrowo, C o n d o ; and Amur Mitiche, INRS-EM?: Canada
SUPER-RESOLUTION WITH SIGNIFICANT ILLUMINATION CHANGE ................................ 1771
Wen-K Zhao, Samoff Corporation, USA
ZOOM BASED SUPER-RESOLUTION THROUGH SAR MODEL FI'ITING . . . . . . . . . . . . . . . . 1775
M. I!Joshi and Subhasis Chaudhuri, /IT Mumbai, Indiu

xxii
HIGH-IIIMENSIONAL MUTUAL INFORMATION ESTIMATION FOR IMAGE REGISTRATION . . . . . . . . 1779
Jan Kybic, Center for Machine Perceplion, CTU, Czech Republic
IMAGE SCALE AND ROTATION FROM THE PHASE-ONLY BISPECTRUM.. .. ............................... 1783
Oulu. Finlund
Jnnne Heikkil'rr, University of
LIGHTFIELDCOMPLETION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Limn Ycttziv, Guillermo Supiro, University of Minnesotn, [JSA; ond Murc Levox Strindford University, USA
ACCELERATION OF CSRBF-BASED IMAGE RECONSTRUCTION B Y WAVELET DOMAIN PRECONDITIONING
Luis Diago, CUJAE University, Cubu; Musuki Kitago and Ichiro Hagiwaru, Togo Institute of Tech~olog?,Jupm
MOTION ESTIMATION IN HIGH RESOLUTION IMAGE RECONSTRUmION FROM COMPRESSED VIDEO SEQUENCES . . 1795
Luis D. Aliwez. Rafoel Moiinu, Liniversihl of Granoda. Spain; oncl Aggelos K. Kut.srrggelus, Nortlrwestern Univer.siy, USA

SESSION TP-LZ: Deblocking, Restoration, and Enhancement

DERlNClNG AND IIEBLOCKING DCT COMPRESSION ARTIFACTS WITH EFFICIENT SHIFTED TRANSFORMS . . . . . . . . 1799
Ramin Su~~~odnni, H P Lnbs, Pnlo Alto, USA; Arvind Sundnmrujm, Xilim mrpomtion, USA; mid Amir . S d , HP ILibs, Pulu Aim. IJSA
ADAPTIVE FUZZY POST-FILTERING FOR HIGHLY COMPRESSED VIDEO.. . . . . . . . . . . . . . . . . . . . 1803
Huo-Song Kong, Mits~bishiElectric Research Lubomrories, USA; Yno Nie, University of Deluwrrrc, USA; Anthony Vetrw, Hoifrmg Sun;
Mitrubislzi I h c t r i c Re,ser,rck I ~ i b s USA;
, und K W " h Bnrnez lJniiw~it?of Deluw0rc, lJSA
NONLINEAR WARPING FUNCTION RECOVERY BY SCAN-LINE SEARCH USING DYNAMIC PROGRAMMING . . . . . . . . 1807
Futih Porikli, MERL,, USA
BAYESIAN POSTPROCESSING ALGORITHM FOR DWT-BASED COMPRESSED IMAGE ......... 1811
Wci Wen,Zlri?un Xiuo u r d Silong Peng, Nririonul ASIC Design Engineering Cenrer. b7stitlrte of Automntion. Chinrse
Sciences, Cliina
OPTIMAL SPARSE REPRESENTATIONSFOR BLIND SOURCE SEPARATION AND BLIND DECONVOLUTION:
ALEARNINGAPPROACH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1815
h!iclclroel Rronslein, Alexnnder Bronslein, Micho~lZibrrlevskv and Yehoshuo Zee
PIXEL RECOVERY VIA LI MINIMIZATION IN THE WAVELET DOMAIN ......... 1x19
lvun Selesnick, Richard I/rn Slyke, Polytechnic Uniwersir): USA; rind Onur Girl
AUTOMATED TREATMENT OF FILM TEAR IN DEGRADED ARCHIVED MEDIA.. .................................. 1823
David Currigan and Ani1 Kokamm, Drpartment of Electrical and EIectmnic Engineering, Trinify College Dublin, Ireland
CORRECTION OF UNDEREXPOSED IMAGES USING SCENE RADIANCE ESTIMATION ............................. 1827
Antonio Robles-Kelly and Edwin Hancock. Tlze Universip of York, UK

SESSION TP-W: Motion Estimation and Detection

AN ADARIVE SCHEME FOR ESTIMATING MOTION ..
Hursan Foroosh. University of C6nfral Florida, USA
ROBUST OFTICAL FLOW FROM PHOTOMETRIC INVARIANTS ..............
Joost Van de Weijer and Theo Gevers, University of Amsterdam, The Netherlands
AN ACCURATE AND ADAPTIVE OFTICAL FLOW ESTIMATION ALGORITHM. ..................................... 1839
Chin-Hung Teng, Depurtment of Electrical Engineering, Narional Tsing Hua University, Taiwan; Shnng-Hung h i , Department of
Computer Science, National "sing Hua University, Taiwan; Yung-Sheng Chen, Depurtmenf of Electrical Engineering. Yuan Ze University,
Taiwun; and Wen-Hsing Hsu, Department of Eleectricul Engineering, National Tsing H u University, Taiwan
EVOLUTIONARY AGENTS FOR EPIPOLAR GEOMETRY ESTIMATION . . . . . . ... 1843
Miflgxing Hu, Virtual Engineering Centre, Queen k University Be!fast, UK; Gordon Dodds, School of Electricul & Electmnic
Enginerring, Queen S Universiry Belfast, UK; and Buozong Yuan, lnsrirute of lnfonnution Science, Beijing Jiaotong University, China
ADAPTIVE EIGEN-BACKGROUNDS FOR OBlECT DETECTION .................................................. 1847
Jonathan Rymel, John-Paul Renno, Darrel Greenhill, Digital Imaging Research Centre, UK; James Onvell and Graeme A. Jones,
Kingston University, UK
CYLINDRICAL SENSOR CALIBRATION USING LINES ....... ...... ...... 1851
Luurenr Smudja. Ryad Benosman and Jean Devars. LlSlE F
MOTION ESTIMATION FROM MOTION SMEAR - A SYSTEM IDENTIFICATION APPROACH ......................... 1855
Om Ji Omer; Sameer Kuinas Indian lnsrirute ofTechnolog?i, Kanpu': India; Rujeev Bojpai, Emuzed India, India; K. S. Venkutesh ond
Sumana Gupta, Indian Institute of Technology, Kunpur, India
ANALYSIS OF IMPULSETRAIN ILLUMINATED IMAGES FOR 2D VELOCITY MEASUREMENT ..... . . . . . . . 1859
Jan Horn, lnstitur fuer Mess- und Regelungstechnik, Universitaet Karlsruhe (THJ. Germany

SESSION TP-L4: Image Segmentation

STATISTICAL SOLUTION TO WATERSHED OVER-SEGMENTATION,.............................................. 1863
Valentin Gies and Thierv M. Bernard. ENSTA LEI, France

xxiii
SEGMENTATION OF REMOTE-SENSING IMAGES BY SUPERVISED TS-MRF. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1867
Giovanni Poggi, Giuseppe Scarpa. Universir'o di Napoli Federico 11, lrrdy; and Josirrne Zerubia, INRIA Sophia Anripolis, France

~~~ ~~~~~~~~ 1x71

Aly Farag. Ayman El-Boz, UniversiQ of Louisville-, USA; rind Grorgy Gimel'farb, Univerriry of Auckland, New Zealand
DEFORMABLE STRUCTURAL MODELS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1875
Sreven Bergrze,: Simon Fmsrr L'nivrrsir?: Canada: Srepha~inAI-Zubi and Klaus 7"bnnies. Orlo-von-Guerickc U n i o ~ r s iGermany
~,
GAP CLOSURE IN (ROAD) NETWORKS USING HIGHER-ORDER ACTIVE CONTOURS . . . . . . . . . . . . . . . . . . . . I879
Marie Rochen: Ian J e n n y and Josiune Zerubia, Arianu (joint reseun:lr group CNRS/lNRIA/UNSA), France
SHAPE ESTIMATION OF 3-D DNA MOLECULES FROM STEREO CRYO-ELECTRON MICRO-GRAPHS . . . . . . . . . . . . . . . . . 1883
Marhews Jacob, Beckinan Insrirure, UIUC. USA: Tlrier? Blu and Michael Unsec EPFL, Switzerland
CONTEXT-DEPENDENT TREE-STRUCTURED IMAGE CLASSIFICATION USING THE QDA DISTORTION MEASURE
ANDTHE HIDDEN MARKOV MODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I887
/(ivanc Ozonar and SangHo Yoon, Sranford Universir): USA
REFERENCE LINE APPROACH FOR VECTOR DATA COMPRESSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1891
Ale.rander AkLnov. Alrrrmd~rKolesnikov mid Pas; Frcmri, Universiry of Joensuic. Finland

SESSION TP-LS: Biomedical Image Processing: Compression, Registration and Fusion

GENEKALIZED SUBSPACE RULES FOR ON-LINE PCA AND THEIR APPLICATION IN SIGNAL A N D
IMAGE COMPRESSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1895
Toshilira Tannka, Tokyo U,rioe,rit? of Agriculrm
A NOVEL PROGRESSIVE THICK SLAB PARADIGM FOR VOLUMETRIC MEDICAL IMAGE COMPRESSION
AND NAVIGATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1899
Bhomrlr Kumar SI: SudipraM~,khopnd/~y',ynsd Visliratn Nandrdka,: Gniernl Elecrric-Global Research Bongalore, India
WAVELET DIFFERENCE REDUCTION WITH REGION-OF-INTEREST PRlOKlTY IN MULTISPECTRAL
VIDEO - SMALL TARGET DETECTION .................... 1903
Yee Lnn! hivemiry of Cal$ornia, San Diego. enrec Dahlgren Division,
USA; and Trunng Nguven. Universiry of CaliJiomia, San Diego, USA
TOWARDS A NEW DIAGNOSIS AID OF CARDIOVASCULAR DISEASES USING 2D-MULTIMODAL
DATA REGISTRATION AND 3D-DATA SUPERIMPOSITION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1907
Gaelle Valet, Sr4phane Sanchez. Juan Manuel Lopez-Hemandez, Chrisrian D a d , Didier WO& CRAN - INPL - ENSEM, France; and
Gilles Karclrec C H U Nancy, France
TOWARDS A FRACTIONED TREATMENT I N CONFORMAL RADIOTHERAPY USING 3DMULTIMODAL
DATAREGISTRATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1911
Ruben Pusoda. Chrisrian Dad, Didier Wolf, CRAN-ENSEM-INPL France: Pierre Alerri. Centre A l a i s Vaurnn. France: and Roseber
Miranda, CRAN-ENSEM-INPL, France
SUB-PIXEL REGISTRATION AND ESTIMATION OF LOCAL SHIFTS DIRECTLY IN THE FOURIER DOMAIN.. . . . . . . . . . . 1915
and Murar Balci, University of Central Florida, USA
Hassan F o x "
FAST ALIGNMENT OF DIGITAL IMAGES USING A LOWER BOUND ON AN ENTROPY METRIC.. 1919
Mer? Sabuncu and Perer Ramadge, Princeton University, USA
INTEGRATED REGISTRATION OF DYNAMIC RENAL PERFUSION MR IMAGES, ................................... 1923
Ying Sun, Camegie Mellon Universiry. USA; Marie-Pierre Jolly, Siemens Corporare Research, USA; and JosC Moura, Carnegie Mellon
University, USA

SESSION TP-P1: Stereoscopic and 3-D Processing I1

3D ORTHOGRAPHIC RECONSTRUCTION BASED ON ROBUST FACTORIZATION METHOD WITH OUTLIERS.. ........ 1927
Li Xi, lnrrirure ofArrijicia1 lnrelligence and Roborics, Xinn Jiaorong U n i v e r s i ~Xian,
, Shaanri Province 710049, PR China, China
SURFACE SEGMENTATION USING A MODIFIED BALL-PIVOTING ALGORITHM ..... ........ 1931
Andres Resrrepo Spechr and Michel Devy, LAAS-CNRS, France
A GENERAL FRAMEWORK FOR 3D SOCCER BALL ESTIMATION AND TRACKING. ................................ 1935
Jinchang Ren, James Omell, Graeme Jones and Ming Xu. Digirol l m g i n g Research Cenrre, Kingston University, UK
MULTISCALE SURFACE REPRESENTATION AND RENDERING FOR POINT CLOUDS 1939
Sung-Bum Park Satzg.Uk Lee, Seoul Narional University, Sourk Korea; and Hyeokho Choi, Ri
AREA MATCHING BASED ON BELIEF PROPAGATION WITH APPLICATIONS TO FACE MODELING ................... 1943
Davide Onofrio, Augusro Sarti and Srefano Tubam, Polirecnico di Milano, lraly
OPTIMIZED SPACE SAMPLING FOR CIRCULAR IMAGE CUBE TRAJECTORY ANALYSIS ................... 1947
lngo Feldmann. Perer Kauff and Perer Eiserr, Fraunhofer lnsrirure for Telecommunicarions,Heinri Insrirure, Germany
PERFORMANCE ANALYSIS OF AN IMPROVED TENSOR BASED CORRESPONDENCE ALGORITHM
FOR AUTOMATIC 3D MODELING . . . . . . . . . . . . . . . . 1951
A j m l Mian, Mohammed Bennamoun

xxiv
A STRUCTURE-FROM-MOTION METHOD FOR 3-D RECONSTRUCTION OF MOVING OBJECTS FROM
MULTIPLE-VIEW IMAGE SEQUENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I955
Ha Le, Dupnrrmenr of Electrical ond Cornpurer Engineering, Viemom Narional Univerriry, Hunoi, Viemam
A FACTORIZATION-BASED PROJECTIVE RECONSTRUCTION ALGORITHM WITH CIRCULAR MOTION CONSTRAINT . . 1959
Yon ti, Wai-Kai Tang rind Yeung-Sum Hung, Drportment of E/eclrical and Elecmnmic Engineering, Tire University of Horrg Kong, Hong Kong
3D GAIT ESTIMATION FROM MONOSCOPIC VIDEO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1963
Angel D. Suppa, CVC, S p i n : Niki Aifmri, Soririr Mrduss
MULTI-RESOLUTION STREAMING AND RENDERING OF 3-D DYNAMIC DATA.. .................................. 1967
Ruinanarhan Submmoniurr, Ashruf Kussinr, Nationol UniversiQ of Singapow. Singupore; Sumit Gupm, Indian Insfititre of Technology,
Roorkee. India; and Kuntal Senyupra, Narionul Universig of Singapore, Singapore
MULTI-RESOLUTION SURFACE RECONSTRUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 I
Mingw He, Bangshu Xiong and Huajing Yu. Norrhwestem Pol~terl,nicalUniversiQ, Clrinrr
PROGRESSIVE COMPRESSION OF 3D DYNAMIC MESH SEQUENCES ...................................... 197s
Jeong-Hw Yung, S<wrd Nm'l Univ., Sourlt Koren; Chnng-Su Kim, Tlrc Clrinrse Unia of Hong Kong. Hong Kong: und Snn,c.Uk Lee, Seoul
Nat '1 Unit,., Sorrtlr Korea
IMAGE FUSION AND WAVELET ANALYSIS FOR 3-D RECONSTRUCTION USING 2-D IMAGES
OBTAINED UNDER DIFFERENT ILLUMINATION CONDITIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1979
Tlrrimas Heget: M~,rllirrko,-Pmdir ond Roher? Mei.wer; rJnivar.siry of Krrise,slr,iireni. G~rmnny

SESSION TP-PZ: Face Detection, Recognition and Classification Ill

IMPROVING INDOOR AND OUTDOOR FACE RECOGNITION USING UNIFIEI) SUBSPACE ANALYSIS
ANDGABOR FEATURES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19x3
Xinognng Wnrig mid Xiuoou Ping, The Cltinme Univcrsiry of Hong Kong, Hong Kong
A FACE IMAGE RECOGNITION SCHEME WITH STRONG TOLERANCES TO LIGHTING FLUCTUATIONS. . . . . . . . . . . . . . 1987
Kenji Marruo. Mrisuyuki Husliiriroro ond Arsushi Koike. KDDl R&D Idm-arories Inc., Japan
SEPARATING USEFUL FROM USELESS IMAGE VARIATION FOR FACE RECOGNITION ............................. 1991
Peter Kahmiii. UCL4, USA
ROBUST NOSE DETECTION IN 3D FACIAL DATA USING LOCAL CHARACTERISTICS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1995
Chenglrun Xu, Yunkong W m g , Tieniu Tan, Narional Lnb of Partern Recognirion, Chino; and Long Quan. Depamaent of Computer
Science, HKUSZ China
FACIAL SIMILARITY ACROSS AGE,DISGUISE, ILLUMINATION AND POS .................................. 1999
Nuravanan Rarnanarhan. Elecrrical & Cornpurer Engineering, USA; Amit Chowd iversity of California, Riverside, USA; and
Rumu Chellappa, Centre f o r Automarion Research, USA
3D FACE RECOGNITION USING LOCAL SHAPE MAP.. .......................................................... 2003
Yueming Wang, Gang Pun and Zhaohui Wu,Zhejiang Universiq, China
COLOR CHANNEL ENCODING WITH NMF FOR FACE RECOGNITION ........................................... ,2007
Menuku Rujapokse, 12R. Si&pore; and Jagarlt Rajapahe, NTU,Singopore
ANOVELCOMBINED FISHERFACES FRAMEWORK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2011
Wenshu Li, College of Conrpurer Science and Technology. Zhejiang Universiiy, China; and Changle Zhou, Deparrmmr of Anijiciul
Intelligence,Xiamen Universiry. China
FACE DETECTION IN THE COMPRESSED DOMAIN ............................ 2015
Pedm Fonseca. Incriruio Superior Tecnico - Inctirufode Tel udba, Philips Research, The Nellierlands
ONLINE FACE RECOGNITION SYSTEMFOR VIDEOS BASED ON MODIFIED PROBABILISTIC NEURAL NETWORKS
Jun Fun, Nevenka Dimirrova, Philips Research USA, USA; and Vasunth Plzilomin, Philips Research Aadren, Germany
DISCRIMINATIVE LIP-MOTION FEATURES FOR BIOMETRIC SPEAKER IDENTIFICATION ......................... ,2023
H. Ertan Ceringul. Yucel Yemez, Engin Erzin and A. Mural Tekalp, Koc University, Turkey
FEATURE-BASED DETECTION OF FACES IN COLOR IMAGES .................................................. ,2027
Huung Chi-Jaung, Ho Ming-Che and Chiang Cheng-Chin. Nurional Dong Hwa Universiiy, China
PATTERN RECOGNITION IN COMPRESSED DCT DOMAIN 203 1
Daidi Zhong and lrek Defe, Tampere Universiry of Technology,Fin

SESSION TP-P3: Video Streaming and Networking

OPllMAL RETRANSMISSION TIMEOUT SELECTION FOR DELAY-CONSTRAINED MULTIMEDIA COMMUNICATIONS. . ,2035
Jiunfei Cui, Wenning Zhun, Nanyang Technological Universiiy, Singapore; and Zhihui He, Universiry of Missouri- Columbia, USA
PEER-TO-PEER MULTIPOINT VIDEOCONFERENCING ................................ ,2039
M. Relw Civanlur; Omur Ozkasnp and Tahir Celeb;, Koc University, Turkey
OPTIMAL RATE AND INPUT FORMAT CONTROL FOR CONTENT AND CONTEXT ADAPTIVE VIDEO STREAMING . . . ,2043
Tanir Ozcelebi, M u m Tekalp and Reha Civanlar; Koc University, Turkey

xxv
 CROSS LAYER OWIMIZATION FOR WIRELESS MULTI-USER VIDEO STREAMING. ............................... ,2047
Eckekard Steinbacli. h i -U Cltoi, Mtmicli UniversirJ.of Technolog): Gernirvg: and Wolfgang Kellerrc DoCoMo C o , n s ! w k " n
hhorarories Europe QnbH, Germany
RATE-DISTORTION-COMPLEXITYADAPTIVE VIDEO COMPRESSION AND STREAMING .................... .2051
Mihacla Van der Sclmac UC Davis, USA; Deepak Turaga, Son? Elecrronics. USA; and Venkmsh A
LOW-COMPLEXITY RATE-DISTORTION OPTIMIZED VIDEO STREAMING. . . . . . . . . . . . . . . . . . . . : . . . . . . . . . . . . . 2055
Jacob Clinkureski, Srunford UniversirJ.,USA; John Aposrolopoulos. Hewlerr-Packnrrl laborarories, USA; md B u n d Girod, Stmford
Universiry, USA
EFFICIENTCONGESTION CONTROL FOR STREAMING SCALABLE VIDEO OVER UNRELIABLE IP NETWORKS . . . . . ,2059
Palrick Srcsnngn and Mrimil Mmdal, Universip of Al6erra. Connrln
IMPROVED MACROBLOCK-BASED REVERSE PLAY ALGORITHM FOR MPEG VIDEO STREAMING. . . . . . . . . . . . . . . . . ,2063
Clirrizg-Hung Fu, Yui-Dm Clian and Wcin-Chi Sio, The Hong Kong Pol.yreclmic Univemiq, Hong Kong
A REGION BASED MULTIPLE FRAME-RATETRADEOFF OF VIDEO STREAMING.. . . . . . . . . . . . . . . . . . . . 2067
Wei Lni. Uniiwrsiry of Science und Teclinolugy of China, China; Xioodong Gu. Micromfr Rrseat RenHua Wong, I.iRong
Iloi, Univrrsiry of Science and Teclstologpof China. Chino; ond Hongjiang Zlmng, Micromft Research Asin, China
FLEXIBLE P-PICTURE (FLEXP) CODING FOR THE EFFICIENT FINE-GRAULAR SCALABILITY (FGS). . . . . . . . . . . . . . . . .,2071
You Z/lori, Uit~iwxityof Scienct. d; Technolog? of Beijing. China; X i a o w n Sun, F e y Wu. MicmwJ~Resea.rrrchAsio, Cllino; HonR Bao.
Ui~ivcr.siry<f.%ie,rce d; Technology of Beijing, Clrino; orid Shipmg Li, Micmsofr Rrseo.n,r.lrAsin. Chino
INCREASINGBANDWIDTH UTILIZATION I N NEXT GENERATION IPTV NETWORKS.. ............................ ,207s
Ulf Jennrlfrrg orid 7ingring Zliang, Mi&S~rsedmUnharsip, Sweden
CHANNEL-AWARE RATE-DISTORTION OPTIMIZED LEAKY MOTION PREDICTION . . . . . . . . . . . . . . . . . . . . . . . . . . .
Zhen 1.i and Edward Dclp. School of Elrcrrical and Compurer Engineering, Pardtre Universit?, West Lnfr,,me, USA
DISTORTION-BUFFER OPTIMIZED TCP VIDEO STREAMING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2083
Anshril Selfgal. Universirv of Illinois, Ur6ana, USA; Olivier Verccheidre.IBM 7: J. Watson Rcs~arch.USA; and Pascal Frossrrrd. EPFL,
Swirzerlund. Swirzerlonrl
OPTIMAL PER-PIXEL ESTIMATION FOR SCALABLE VIDEO CODING. . . . . . . . . . . . . . . . . . . . . . . ,2087
Arlranasios Leonrarir and Pamela C. Cussam,OnisersirJ.of Califomin, San Diego, USA

SESSION TP-P4: Shape Extraction and Analysis

COMPLEX CURVE TRACING BASED ON A MINIMUM SPANNING TREE MODEL AND REGULARIZED
FUZZY CLUSTERING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2091
Lorn Shu Yan and Yan Hong, Depamnent of Conipurer Engineering and Informarion Technolog)
SHAPE CLASSIFICATION OF PARTIALLY OCCLUDED OBJECTS USING SUBSPACE DETECTORS . . . . . . . . . . . . . . . . . . . ,2095
Arnr B. Salberg. AlfHarbirz, Instirure of Murine Research, Dept. Tromsm, Noway: and Alfred Hunssen, Univsersir) of T r o m s ~Depf.
Physics,Nowrry
THREE-DIMENSIONAL INTRINSIC SHAPE. . . . . . . . . ...................... ,2099
Victor Ha, Samsung Informarion Sysrems America, USA; a
A ROBUST SHAPE-FROM-SHADING USING MULTIPLE SURFACE NORMAL APPROXIMATIONS. ................... ,2103
Osamu Ikeda, Takushoku Universir), Japan
SURFACE HEIGHT RECOVERY FROM SURFACE NORMALS USING MANIFOLD EMBEDDING. ..................... ,2107
Antonio Robles-Kelly and Edwin Hancock, The University of York. U K
ROBUST CIRCLE DETECTION USING A WEIGHTED MSE ESTIMATOR. . . . . . . . ..2111
Guido Schusrer, Hocl~schuleRapperswil, Swirzerland; and Aggelos Karsaggelos. Norrhwesrern Universily, USA
REDUCED CROSS-DISCRIMINATION FOR DISCRIMINATIVE FILTERS ...................................... 211s
Alexandre Mendonc,a, Insriruro Militar de Engenharia - Dm3, Brazil; and Silva, Universidade Federal do Rio de Janeiro -
Programa de Engenharia ElirricdCOPPE, Brazil
EXTRACITNG PROJECTIVE INVARIANT OF 3D LINE SET FROM A SINGLE VIEW ................. 2119
Zhe Chen, School of Compurer Science,Norrhwestem Polyrechnical University, China
A NEW REPRESENTATION OF CHARACTER SHAPE AND ITS USE IN ON-LINE CHARACTER RECOGNITION
BYASELFORGANIZINGMAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2123
Neila Mezghani, Amor Miriche, INRS-EMZ Canada; and Mohamed Cherier, ETS, Cnnadu
CONSTRUCTING THE TOPOLOGICAL SOLUTION OF JIGSAW PUZZLES ............. 2127
Johan De Bock, Parrick De Smer. Wilfried Philips.and J o h n D'Haeyec Ghenr University, Dep.
BLURRED IMAGE RECOGNITION BASED ON COMPLEX MOMENT INVARIANTS. ................................ .2131
Zhang Tianru and Liu Jin, HuaZhong University of Science and technology, China
NEW FEATURES FOR AFFINE-INVARIANTSHAPE CLASSIFICATION ............................................ ,2135
Carlos Ramon Panraleon Dionisio and Hae Yong Kim, Universidade de Sm Paulo, Escolo Poliremica, Brazil
WAVELET APPROXIMATION-BASEDAFFINE INVARIANT 2-D SHAPE MATCHING AND CLASSIFICATION 2139
lbrahim El Rube', Molramed Kame/, Universir) of Warerloo, Canada; and Maher Ahmed, Wilfrid OIurier University, Canadu

xxvi
SESSION TP-P5: W a t e r m a r k i n g IV
DIGITAL WATERMARKING BASED ON LOCALLY LINEAR EMBEDDING. . . . . . . . . . . . . . . . . . . . . . . . . . . ,2143
Yonggang Fu, Rubnin .%n and Liping SIien, Slranglrai Jirrorong Uniwrsit?: Clrina
APPLICATION OF BPCS STEGANOGRAPHY TO WAVELET COMPRESSED VIDEO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2147
Hideki Nodu, Tumonori Furura, Michiharu Niirni nnd Eiji Kawaguchi, Kvushu Iilsrirrrre of Technology,Japm
REDUNDANT IMAGE REPRESENTATIONS IN SECURITY APPLICATIONS ................................ .2151
Philippe Just, Pierre Vandergheynsr and Pascal Fmssurd, Swiss Federal lnsrirure of Technology, Switzerland
A STUDY ON A WATERMARKING METHOD FOR BOTH COPYRIGHT PROTECTION AND TAMPER DETECTION.. . . . . .2155
Jun Walmribe, hidoko Hasegawa and Shigeo Kato. Faculry oJ Engineering, Ursunorrziyn Univer.~its,Japrm
BLIND MPEG-2 VIDEO WATERMARKING ROBUST AGAINST SCALING.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ndin Whng and Alan Peurrnoin, Queen Ma7 College, Universir? of landon, UK
MULTIPLE-I)ESCRIFTION CODING FOR ROBUST IMAGE WATERMARKING .............. ,2163
Eng-Fen H.Tkr. Clien-Yko Clrrrng m d Jay-Rciy Liao, Drp-menr rJ'EIecrrica1 Enginee"ng, ,iw~<in
SHORT N-SECURE FINGERPRINTING CODE FOR IMAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2167
Won-g?rori Kim, ETRI, South Koreo; and Yvrrnglzo Sah, ETRI/Digital Cnnrenrr Research Division, Soatlr Koren
NEW REAL-TIME WATERMARKING ALGORITHM FOR COMPRESSED VIDEO IN VLC DOMAIN .................... 2171
Hefei Ling, Zlrmgding O r ond Ftrhrio Zou, Scliool oJ Compurer Scirnce,Huo:hong Uniaerrity of.%iencect Technolog? (HUSTJ, Clrina
ROTATION AND CROPPING RESILIENT DATA HIDING WITH ZERNIKE MOMENTS. . . . . . . . . . . . . . . .2175
P d d K. Aniirz nnd Koddavawr P S~rbholakslsni,Srevens Insrirule of Technolop\: USA
NON-UNIFORM QUANTIZER DESIGN FOR IMAGE DATA HIDING.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2179
Ning 1,iu rind K.P. Suhbnlak.dt,ni. st eve,^,^ 1n.sriture oJTeclmolog?, USA
COPYRIGHTTRACING USING INVARIANTS OF CONTENTS ....................... ..2183
Jinhrci Cliao, Slrintum Suzuki and Jongdne Kim, Clrico Universilx J q a n
OF'TIMAL WATERMARK DETECTION ON INTERPOLATED IMAGES UNDER NOISE.. . . . . . . . . . . . . . . . . ..2187
Alexia Giannorrkr. Nikolaos V Boulgouris and Diirtirrios Hnrzinnkos, L'niuersirv of Tomnro, Gmnda
WATERMARKING COLOR HISTOGRAMS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 I9 I
Sujov Roy nnd Ee-Clrien Cltang, NUS, Singapore
DH-LZW LOSSLESS DATA HIDING IN LZW COMPRESSION.. .. ............... ,2195
Hiuk Jue Sliim, Jinlweng Ahn and Byeungwoo Jean, Sungkyunkwan University. Korea

SESSION TP-P6: I m a g d v i d e o Storage a n d Retrieval

TOWARD AN IMPROVED ERROR METRIC.. ........ ................................................ ,2199
Qi Tim. Qing Xue, Jie Yu, University of Texas ar San Anronio, USA; Nicu Sehe. University oJAmsrerdam, The Nerher1and.r; and Thomas
S.H u n g , University ~ J l l l i n o i sat Urbana-Chanipaign,, USA
RELATING WORDS AND IMAGE SEGMENTS ON MULTIPLE LAYERS FOR EFFECTIVE BROWSING AND RETRIEVAL.. 2203
Andrea Kutics. Tokjo University oJTeclmology / N T T D a l a Corporation, Japan; Akihiko Nnkagawa, N7T Dara Corporation, Japan; Slioji
Ami, Japan Sysretns Co., Ltd., Japan; Himyuki Tanaka and Sakuichi Okrsuka, N1T Data Corporation, Japan
MULTI-LABEL SVM ACTIVE LEARNING FOR IMAGE CLASSIFICATION ......................................... ,2207
Xuchun Li, Lei Wang and Eric Sung, Nanyang Technological University, Singapore
GENERALIZATION OF THE ANGULAR RADIAL TRANSFORM .................................................. .221 I
Julien Ricard, David Coeurjolly and Arilla Baskurt, LIKIS, France
MULTI-LAYER SEMANTIC REPRESENTATION LEARNING FOR IMAGE RETRIEVAL ...... ............ ,2215
Wei Jiang. Guilzua Er and Qionghai Dai, Deparrmenr oJAuromarion, Tsinghua University, China
RETIN A L AN A C n V E LEARNING STRATEGY FOR IMAGE CATEGORY RETRIEVAL ...................
Philippe-H Cusselin and Marthieu C o d , ETIS CNRS UMR 8051, France
ACTION SEGMENTATION AND RECOGNITION IN MEETING ROOM SCENARIOS.. :. . . . . . . . . . . . . . . ,2223
Frank WdlhofL Martin 2061 and Gerhard Rigoll, Technical University of Munich, Germany
SHARING VIDEO ANNOTATIONS ........................................................................ ,2227
Yamn Caspi, The Hebrew Universiry of Jerusalem, Israel; and David Bargeron, MicmsoJi Research, USA
COMBINING LUMINANCE AND EDGE BASED METRICS FOR ROBUST TEMPORAL VIDEO SEGMENTATION . . ,2231
Edmundo Suez Jose Benavides, Universiry of Cordoba, Spain; and Nicolas Guil, University oJMalaga, Spain
A NOVEL COMPRESSED DOMAIN SHOT SEGMENTATION ALGORITHM ON H.264lAVC ........................... .2235
Yang Liu, Wei Zeng. Harbin hstirure oJTechnology, China; Weiqiang Wang, Groduare College. Chinese Academy of Science, China; and
Wen Goo, Hnrbin Institute of Technology. China
A NOVEL HASHING ALGORITHM FOR VIDEO SEQUENCES.. ............................................ ;. . . . . .2239
Antonio Mucedem, Rosa Lancini and Francesco Mapelli, CEFKIEL .Polirecnico di Milnno, Italy
TRAJECTORY-BASEDVIDEO RETRIEVAL BY STRING MATCHING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2243
Jun Wei Hsieh, Shang-Li Yu and Yung-Sheng Chen. Department oJElecrrica1 Engineering, Yuan Ze University, Taiwan

xxvii
NEWS SPORTS VIDEO SHOT CLASSIFICATION WITH SPORTS PLAY FIELD AND MOTION FEATURES 2247
De-Hong Wang. a i Ean. Slieng Gno, lnsrirurc for hlfoco,wn Research, Singapore: and Wing-Kin Sung, Nariunal Univr.r.sily ,$Singapore,
Singapore

SESSION TP-W: Wavelet Video Coding and Scalability I1

ADAPTIVE LOSSLESS VIDEO COMPRESSION USING AN INTEGER WAVELET TRANSFORM.. . . . . . . . . .
Sohng-Gyr Park, Edward Ddp, Sclrool of Elecrrical and Computcr Engineering, Furdue Univer.Yi0. USA: and Haop
Corporarim, lndianapoli.s, IN, USA
A MODEL-BASED MOTION COMPENSATED VIDEO CODER WITH JPEG2000 COMPATIBILITY ..................... .225.5
Marco Cagnozzo. Tlzomas And1-4, Marc Anronini and M i d e l Barlaud, 13s Labomroc, UMR 6070 of CNRS, Univerrir? of Nice-Sophia
Anriplis, Frrrncr
DRIFT REDUCTION FOR A H.264lAVC FINE GRAIN SCALABILITY WITH MOTION COMPENSATION ARCHITECTURE . . .2259
JoOo Ascmso, Essrola Superior de Trcnologia de Serdbal, bzsrirrrro dos TL.Ieco,,,rmicacor.s.Porrugal: and Fernando Pereim, Insriruro
Superior Tdcnica Insriruro dos T~lecoiiainicocoes.Porrugal
MULTIPLE DESCRIPTION WAVELETCODING OF LAYERED VIDEO . . . .2263
Nikolaos Boulgorwi.c. Uniwr.siry of Toronro, Canada: Konsronrinos Zoclzariadi.~,Norrlnwsrem Univerir?, USA: Angelvs Kanlis and
Micfiael Srrinrzis, I~~for~~raric.~
and Teleniarics Insrirure, Greece
ENERGY DISTRIBUTED UPDATE STEPS(EDU1 IN LIFTING BASED MOTION COMPENSATED VIDEO CODING . . . . . . . .2267
Bo Fens, Tyinglrua Uiiiser.sir?, Cliino: Jizlreng Xii, Fen8 Wii, Microsoff Research A s i ~China: Slriyiung Yrng, Knglrua Univt~rrin:Cltina:
and Sliiprwg Li, Microsoff Rrsem-cAAsia, China
LAYERED MOTION ESTIMATION AND CODING FOR FULLY SCALARLE 3 0 WAVELET VIDEO CODING.. . . . . . . . . . . . .2271
Ruiqin Xiong, Iitsrirure of Coiirprrring Technologl: Chinese Acadeiig of Sciences, Cltinu: Jizkeng X u , Feng Wu, Slripeng Li, MicmYoJi
Rcsearch Asia, China: nrid Yr.@ Zlmng. Microsoff, USA
ENHANCED MOTION ESTIMATION FOR INTERFRAME WAVELET VIDEO CODING . . . . . . . . . . . . . . . . . 2275
Cliia-Yong Tsoi, Hon-Kumg Hsn, Hsiang-Cheli Huang, Hsrieh-Ming Hang, Narional Chiao Tung Uniaersir?, Taiwnn: and Cuo-Zua Wu,
Indusrrial Technohgp Reseairlr Insrirure, Taiwan
PERFECT RECONSTRUCTION DEINTERLACER BANKS FOR FIELD SCALABLE VIDEO COMPRESSION.. ........... .2279
Sltogo Muramarsu, Takuma Islridu and Hisolaw Kikuchi, Niigara Universiry, Japm
MOTION COMPENSATED TEMPORAL FILTERING WITH OtTIMAL TEMPORAL DISTANCE BETWEEN
EACH MOTION COMPENSATION PAIR . . . . . -2283
Yang Liu, Nanvang Technological Univerisy, Singapore; Zliengguo ti,lnsrirute for Infocomm Research, Singapore: and Yeng Clmi Soh,
Nanyang Technological Universiry, Singapore
FAST IN-BAND MOTION ESTIMATION WITH VARIABLE SIZE BLOCK MATCHING. ............................... ,2287
Marco Tagliasacclri, Srefano Tubam, Augicsro Sarri and Davide Maesrroni, Diparrimenro di Elertronica e lnfonnuzione - Politecnico di
Milano. lroly
MOTION-COMPENSATED TEMPORAL FILTERING WITHIN THE H.264lAVC STANDARD ........................... .2291
Emrah Akyol, Mural Tekalp and Relra Civanla~KOC University, Turkey
OPTIMUM BIT ALLOCATION FOR FGS VIDEO CODING, ......... .. 2295
fun Xie, Liung-Tien Chia and Ru-Sung Lee, School of Computer Engineering, Nanyang Technological University, Singapore
SIMPLE AVC-BASED CODECS WITH SPATIAL SCALABILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2299
Rafu Lange. Lukasz Baszak and Marek Damahski. Poznah University of Technology, Poland
.MOTION WAVELET DIFFERENCE REDUCTION (MWDR) VIDEO CODEC ......................................... .2303
Yee Lnw and Tmong Nguyen, University of California, San Diego, USA

SESSION TP-PS Image Modeling

THE SALIENCY GROUPING FIELD 2307
Maurizio Pilu, H P Labs, UK
GLOBAL MODULATORY FACTORS IN DCT-BASED JUST-NOTICEABLE-DISTORTION ESTIMATION. . . . . . . . . . . . . . . . . . 2 3 I I
Zhongkang Lu, W S.Lin, X. K. Yang. E. I? Ong and Susu Yao. lnsrirure for lnfocomm research, Singapore
PERCEPTUAL MODEL BASED DATA EMBEDDING IN MEDICAL IMAGES 2315
Somarendm Dandapar, Opas Churarape and Shankor M. Krishnan, EEE, NTU, Singapore
UTILISING OBJECTIVE PERCEFWAL IMAGE QUALITY METRICS FOR IMPLICIT LINK ADAPTATION. . . . . . . . . . . . . . ,2319
Tubagus Maulana Kusuma, Manora Caldera and Hans-Juurgen Zepemick. WATRI, Ausrralia
SCALE-BASED FORMULATIONS OF STATISTICAL SELF-SIMILARITY IN IMAGES ................................ ,2323
Seungsin Lee and Raghuveer Ruo, Rochester Instilure of Technology. USA
PERFORMANCE ASSESSMENT OF A VISUAL AII'ENTION SYSTEM ENTIRELY BASED ON A HUMAN
VISION MODELING ....................................................... 2327
Olivier Le Meur Thom e Barba, IRCCyN, France: and Dominique Thorenu, Thorn". France
COLOR IMAGE COMPRESSION USING ADAPTIVE COLOR QUANTIZATION. ........................ 2331
Chun-Hsien Chou and Kuo-Cheng Liu, Depomnenr of Elecrrical Engineering, Tarung Universiy, Taiwan

xxviii
TEXTURE CHARACTERIZATIONAND COMPRESSION BASED ON HUMAN PERCEPTION IN THE
JPEGZOOO FRAMEWORK. . . . . . . . . . . . . 2115
tiamol Falrsq: Wesr Virginia University,
ESTIMATION OF THE RADIOMETRIC RESPONSE FUNCTIONS OF A COLOR CAMERA FROM DIFFERENTLY
ILLUMlNATEDlMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2139
Kliurmaz Skofique and Mubarnk Shah, Universiq of Central Florida, USA
IMPROVE ROBUSTNESS OF IMAGE WATERMARKING VIA ADAPTIVE RECEIVING.. ............................. ,2343
Xinnpui Kang, Jiwu Hunng, Sun Yar-Sen University, Chino; and Yun Q. Shi, New Jersey hslilure of techno log^, USA
AN EFFICIENT MOTION DETECTION ALGORITHM BASED ON A STATISTICAL NON PAllAMETRlC NOISE MODEL . . .2347
Alessanrlro Bevilucquo, Luigi Di Srefano and Ale.windro Lnnzo, DEIS-ARCES(Cenrrpqf fieelleiice), Universirv of Bologna, Iralv
LEARNING STRUCTURED DICTIONARIES FOR IMAGE REPRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2351
Gianlucrr Monaci ond Pierre Vanderghrytst, Sigrid Processing lr!rrirure (ITS) - EPN., Swir;erlrind
IMAGE UNDERSTANDING AND SCENEMODELS: A GENERIC FRAMEWORK INTEGRATING DOMAIN
KNOWLEDGE AND GESTALTTHEORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2355
Nicokrs ZluroJJ Bruno Tellez m r l Alillu Bus ranee
STOCHASTIC MODELING OF VOLUME IMAGES WITH A 3-D HIDDEN MARKOV MODEL ......................... ,2159
Jirr Li, Dbirrij Joski orid Jmies Wmg. Pennrdivnia Stare University, USA

SESSION WA-SI: Content Understanding for Home Photograph and Video Management
AUTOMATED RED-EYE DETECTION AND CORRECTION IN DIGITAL PHOTOGRAPHS. ............................ 2361
h i Zlarny, fimj>ny Sun, Mingjing Li and Hoqjirrng Zlraiig, Micnisof! Rescorclr Asia, Clrinrr
IMPROVED BLUE SKY DETECTION USING POLYNOMIAL MODEL F I T . . . . . . . . . . . . . . . . . . . . .: .................... .2367
Andrew Gallagher; Jiebo Luo and Wei Hao, Eusrniun Korluk Coiripan>~, USA
UNIFYING LOCAL AND GLOBAL CONTENT-BASED SIMILARITIES FOR HOME PHOTO RETRIEVAL. . . . . . . . . . . . . . . ..2171
Jm-Hwee Lint. lnstirule for Infocown Research, Singapore; and JCSSPJim, Uniw-siq of New Sourlr Wales, Ausrralkt
OVER-COMPLETE REPRESENTATION AND FUSION FOR SEMANTIC CONCEPT DETECTION . . . . . . . . . . . . . . . . . . . . . . . -2375
Aposrol Nrrrrev, Milind Nophade, Ching-King Lin and John R. Smith, IBM T J Warson Recearch, USA

SESSIONWA-SZ: Pattern Discovery in Real-world Broadcast Video

VIDEO MINING: PATTERN DISCOVERY VS PATTERN RECOGNITION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2379
Ajay Divakaran, Kadir Pekeer;Mirsubislri Elecrric Researcl8 Laboratories, USA; Shih-Fu Chang, Columbia University, USA; Regunnflmn
Rndhnkrishnnn, Mirsubishi Electric Research Lnborntories, USA; and h x i n g Xie, Columbia Universiry, USA
DISCOVERING MEANINGFLIL MULTIMEDIA PATTERNS WITH AUDIO-VISUAL CONCEPTS AND ASSOCIATED TEXT. .. .2383
Lexing Xie, Lyndon Kennedy, Shih-Fu Chang. Dept. of Elecrrical Engineerng, Columbia University, USA; Ajay Divakaran, Huifmg Sun,
MERL, USA: and Ching-Yung Lin, IBM Research, USA
A COMPARISON OF CONTINUOUS VS. DISCRETE IMAGE MODELS FOR PROBABILISTIC IMAGE AND
VIDEO RETRIEVAL , , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2387
Arjen D e Vries and Thijs Wesremeld, Cenrrum voor Wiskunde en lnfonnntica (CWI), The Netherlands
MULTIMODAL INFORMATION FUSION FOR VIDEO CONCEPT DETECTION
Yi Wu, ECE Depr. Universiry of California Sanra Barbara, USA; Cliing-Yung Lin, IB
Edward !l Chang, ECE Depr, University of California Sanra Barbara, USA; and John R. Smirh, I B M T.J. Warson Research Cenrez
Hawthome, USA

SESSIONWA-L1: Image Scanning, Display, and Printing I

EFFICIENT CLASSIFICATION OF SCANNED MEDIA USlNG SPATIAL STATISTICS ................................. ,2395
Gozde Unal. Siemens Corporare Research. USA; Gaurav Sharma, University of Rochesrez USA; and Reiner Eschbach. Xerox
Corporarion, USA
EFFICIENT INSCRIBING OF NOISY RECTANGULAR OBJECTS IN SCANNED IMAGES ............................. ,2399
Cormac Herley, Micmsoft. USA
NON-PHOTOREALISTIC RENDERING FROM MULTIPLE IMAGES 3
Alberro Barresaghi, Cuillenno Sapim, University of Minnesota, USA; Tom
A SMALL-SIZE CHINESE FONT DISPLAY BY PERCEPTION-BASED METHOD. .................................... ,2407
Thou-Ho (Chao-Ha) Chen, Yung-CliuenChiou, Deparfmenr of Elecrronic Engineering, National Kaohsiung University of Applied
Sciences, Taiwan; and Chao-Tan Huang, Deparrmenr of Elecrrical Engineering, Narional Taiwan Universiry, Taiwan
RECOVERING IMAGING DEVICE SENSITIVITIES: A DATA-DRIVEN APPROACH.. ......... 241 1
Paulo Camalho, Am"ancio Sanros and Pedro Marrins, CISUC, Ponugal
A CLASSIFICATION APPROACH TO COLOR DEMOSAICKING 2415
Cindy Kwan and Xiaolin Wu, McMasrer Unive&q, Canada

xxix
ILLUMINATION ESTIMATION BASED ON VALID PIXEL SELECTION IN HIGHLIGHT REGION . . . . . . . . . . . . . . . . . ,2419
Olr-Seol Kwon, Yang-Ho Clio, Yun-Zie Kim ond Yeong-Ho Ha, Sclrool of Elecrrical Engineering and Compraer Science, Kyungpook
Narionnl Universi,y Soralr Korea
DETECTION AND DETERRENCE OF COUNTERFEITING OF VALUABLE DOCUMENTS . . . . . . . . . . . . . . . . . .... ,2423
Corrnac Herle?: Microsofr. USA; Poorvi b r a , GWU, USA: and Slruwn Yang, Cisco. USA

SESSIONWA-L2: Image Formation I

AN EFFICIENT PHASE RETRIEVAL METHOD USING SNAKES FOR IMAGE RECONSTRUCTION . . . . . . . . . . . . . . . . ,2421
Keiko Kondo, Miki Haseyama a17d Hideleu Kirnjima. Sclrool of Engineering, Hokkaido University, Jnpan
FMKl SIGNALMODELING USING LAGUERRE POLYNOMIALS . . . . .........................
Vicror Solo, EECS Depamrtenr, Universir? of Michigan, Ami Arboc USA; Clrrisroplw- Long, Mrirrinos NMR Centec Muss. Gr,ierul
Hospirnl, USA; En7e1yN. Brown, Nrurnscimce Balisrics Risrarcli Loh, D e p a r f n w ~olf Anesrlrrsiu and C,-iricnl Care, Mms. General
Huspirol, USA; Elissa A m i n o ~Moslre Bac Marrinos NMR Cmrec Mass. General Hospiral. USA; and Suprarim Salio. cience Stnristics
Research Lnb. Depurrnrenr ofAnesrh8siri and Criricul Core. Muss. Generril Hospiral, USA
SIMPLIFIED DIGITAL HOLOGRAPHIC RECONSTRUCTION USING STATISTICAL METHODS ....................... .2435
Jeflr-ej’A. Fecrlec Uniwrsiry of Miclzigan, USA; and Saowcipak Sufllrivirnr. Nnrional Elecltunic~ond Computers Teclmolog? Cenrel; Tliuilrr,id
A SET THEORETIC APPROACH TO TARGET DETECTION USING SPECTRAL SIGNATURE STATISTICS.. . . . . . . . . . . . . . ,2439
David Rouse ond Joel Trrrssell, NC Sfore Univo-.sify, USA
KNOWLEDGE-DRIVEN SEGMENTATION OF THE CENTRAL SULCUS FROM HUMAN BRAIN MR IMAGES
Wei Zuo, Qinymao Hu, Aamer Azic. Rionrrdical Imaging Group, BII, Sirigq,ore; Ki.S.ck Loe, School of Comprrii!q, NUS, Sinpopore: mid
Wieslrrw Nowinski, Bio!nedical I m a ~ i i i gGroup, 811, Singopore
A FEASIBILITY STUDY ON A NOVEL METHOD OF VISUAL OBSTACLE DETECTION. ............................. ,2447
Andrrej Slurck, Nonynng Technological University, Scliool of Cornputer Engirieering, Singapore; and Cling Seong Rm,Nrinyang
Z~rhnologicolUniversity, Scliool of Mtduinicnl& Producrion Engineering. Singapore
A PRACTICAL ALGORITHM TO CORRECTGEOMETRICAL DISTORTION OF IMAGE ACQUISITION CAMERAS.. . . . . . .2451
Juan Torres and Jost Manrrcl Mmdnnrlez, Universidod Polirknica de Modrid, Espann
HIGH SPEED PROCESSING OF BIOMEDICAL IMAGES USING PROGRAMMABLE CPU ....................
Jin Young Hong, Huinan Conrpurer lnreracrion Master Pmgrarx, USA; ond May D. Wang, The Walluce H. Corclrer Deparrment of
Biomedical Engineering, Georgio Insrirute of Teclinology and Emory Uniwrsity, USA

SESSIONWA-L3: Stereoscopic and 3-D Coding & Processing

REGION-ORIENTED COMPRESSION OF MULTISPECTRAL IMAGES BY SHAPE-ADAPTIVE WAVELET
TRANSFORM AND SPlH ................. ................ ..........
Marc0 Cagnucco, Giovann uisa Vedoliva, Unive
RATE-DISTORTION ANALYSIS OF RANDOM ACCESS FOR COMPRESSED LIGHT FIELDS
Prusliant Ramanarlran and B e d Gimd, Stanford University, USA
MULTI-VIDEO COMPRESSION IN TEXTURE SPACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2467
Gemot Zieglec Hendrik RA. Lensch, Naveed Ahined, Marcus Magnor and Hans-Perer Seidel. M P I Informufik, Saarbriicken, Germany
VIEW-DEPENDENT NON-UNIFORM SAMPLING FOR,lMAGE-BASED RENDERING ..... ,2471
Cha Zhang and Tsuhan Chen, ECE, Camegie Mellon Univ., USA
A FOCUS MEASURE FOR LIGHT FIELD RENDERING.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2475
Keira Takahashi, The Universiry of Tokyo, Japan: Akira Kubora, Camegie Mellon University, USA; and Takeshi Naemura, The Universiry
of Tokyo, Japan
COMPLETE 3-D MODELS FROM VIDEO: A GLOBAL APPROACH. . . . . .......... ... ,2479
Bruno E. Gonc,alves and Pedm M. Q. Aguiar, ISWISZ Porrugal
ROBUST EGO-MOTION ESTIMATION AND 3D MODEL REFINEMENT USING DEPTH BASED PARALLAX M O D E L . . .. .2483
Amit Agrawal and Roma Chellappa, CFAR. UMD, USA
VARIOUS IMAGE TAKING STRATEGIES FOR 3-D OBJECT MODELING BASED ON MULTIPLE CAMERAS
Joo Kooi Tan, lku Yamaguchi, Seiji Ishikawa, Kyushu Insrirure of Technology,Japan; Shunji Himkawa, Kyushu Univ
Tomonori Tabusa. Yuge National College of Technology, Japan

SESSIONWA-L4: Image Coding I

ROBUST WIRELESS TRANSMISSION OF REGIONS OF INTEREST IN JPEG2000 ................................... .2491
Victor Sanchez. Mrinal Mandal, Depr. of EE; Univ. of Alberta, Canada; and Anup Basu. Depr. of CS; Univ. ofAlberta, Canada
BASIS PICKING FOR MATCHING PURSUITS IMAGE CODING. ...................... ...................
Don Monm. University of Barh, UK
MINIMUM HARDWARE IMPLEMENTATION O F MULTIPLIERS OFTHE L l F n N G WAVELETTRANSFORM.. . . . . . . . . . . .2499
Yoshihide Tonomura. Nagaoka University of Technology,Japan; Somcharr Chokchoitam, T / ” m a a r University, Thailand: and Masahiro
Iwahaslzi, Nagaoka University of Technology,Japan

xxx
A WAVELET-BASED TWO-STAGE NEAR-LOSSLESS CODER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2503
Sekoon Yea and William Pearlman. Rensselmr Polvrechnic Insrirure, USA
THEORETICAL ANALYSIS ON OPTIMUM WORD LENGTH ASSIGNMENT FOR INTEGER DCT ...................... ,2507
Mnsnhiro Iwnhashi, Kohtaro Nahrgowo, Nagaoko Universiq of Trclznolog>:Japan: Somclmrr Chokclmirnm, Tlmvmosnr Univemir),
Thailand: und Yoslrihide 7i"mura, Nagaoka Universiq of Tecknology,Japan
DESIGN OF IIR ORTHOGONAL WAVELET FILTER BANKS USING LIITING SCHEME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,251 I
Xi %hang, Universiry of Elrcrr-Co,,i,rricni~arions,Depamnenr of Infinnation and CorrtmunicarinnEngineering. Jnpon; Wei Wany,
EishinoN Yoshikowa and Yoslzinori Eikei, Nugaoka Universin of Technolog\: Japan
ROTATION OF FOVEATED IMAGE IN THE WAVELET DOMAIN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2515
Hang Yu, Vu-Tlmnh Ngqeen and Le-Chien Clwng. Shoo1 of Cnmpufing, Narionnl Univrrsiq of Singapore. Singapore
LOW COMPLEXITY LOSSLESS VIDEO COMPRESSION.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2519
Yao-Clnoi Fmg. Clmn-Yi Lee. Yin-Ming Wang, Ch~mg-NmgWnng and Ehao Clziang. Nnrional Chiao Tung Unirer.sih, 'hiwan

SESSIONWA-LS: Source-Channel Coding I

WIRELESS TRANSMISSION OF IMAGES USING JPEG2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2523
'Nikokiis Tlmmos, Informatics rmd T2lemorics Insrirute, Greece; Niioloos Borrlguurir, Uniorrisq of Toronto, Canarlu; m d Michael
Srrirrt:i.r. Uniwrsiry of Tl~essolmiki.<;recce
ENCODER AND DECODER OPTIMIZATION FOR SOURCE-CHANNEL PKEIXTION IN ERROR RESILIENT
VIIIEOTRANSMISSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2527
Huo Yrrng and Kenneth Rose, Ebcrrical and Compurer Enginrwing Dcportmenr. Universiq of Califomhi Santa Barhara, USA
A N INTEGRATED JOINT SOURCE-CHANNEL CODING FRAMEWORK FOR VIDEO TRANSMISSION OVER PACKET
LOSSY NETWORKS ................................................ 2531
Fan ZIui, Yijach Eise ggelos Karrnggelna Norrlwesrern Univrrriy, USA
LAYERED UNEQUAL LOSS P R O T E a I O N WITH PRE-INTERLEAVING FOR PROGRESSIVE IMAGE
TRANSMISSION OVER PACKET LOSS CHANNELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2535
Jimfei Cai, Ximgjun Li, Nanyang Trclznologicol Universiq, Singapore; and Clrung Wen Chen, Florida Insrirute of 7eclsiolng): USA
RATE-CONSTRAINEDADAPTIVE FEC FOR VIDEO OVER ERASURE CHANNELS WITH MEMORY . . . . . . . . . . . . . . . . . . ,2539
Shirirlr Koronde and Havder Rudlm, Michigan Srare Universir), USA
JOINT OBJECT-HASED VIDEO ENCODING AND POWER MANAGEMENT FOR ENERGY EFFICIENT
WIRELESS VIDEO COMMUNICATIONS ...................... 2543
Haolrvng Wrmg, Yi/rach Eisenberg. Fan Zmi and Aggelos Karsaggelos. Norrlzwestem Universir), USA
CONGESTION-AWARE RATE ALLOCATION FOR MULTIPATH VIDEO STREAMING OVER AD HOC
WIRELESS NETWORKS ..................................... .... ... ... 2547
Xiaoqing Zhu. Sangeun Han and Bernd Gimd, Stanford University, USA
ERROR RESILIENT MQ CODER AND MAPJPEG 2000 DECODING ............................................... ,2551
Marco Grangetto. Enrico Mugli and Gabriella Olmo, Polirecnico di Torino, lraly

SESSION WA-PI: Motion Detection and Estimation: Optical Flow and Change Detection
DENSE MOTION FIELD ESTIMATION BY 3-D GABOR REPRESENTATION. ....................................... ,2555
Mu Feng and Todd Reed, University of Hawaii at Manoa, USA
ESTIMATE LARGE MOTIONS USING RELIABILITY-BASED DYNAMIC PROGRAMMING. .......................... ,2559
Minglun Gong. Laurenrian Univ., Canada; and Yee-Hong Yang, Univ. of Alberta, Canada
AUDIO-VISUAL FLOW - A VARIATIONALAPPROACH TO MULTI-MODAL FLOW ESTIMATION. .................... ,2563
Raffay Hamid, Aaron Bobick, GVU - College of Computing - Georgia lnstirure of Technology. USA; and Anthony Yeui, School of
Elecrriral and Computer Engineering - Georgia Institute of Technology. USA
BENEF'ITS OF TEMPORAL OVERSAMPLING IN OPTICAL FLOW ESTIMATION. ................................... ,2567
SukHwun Lim, John Aposrolopoulos, Hewlert-Packard Laboratories, USA; and Abbas El G a m / , Sranford University, USA
ATLAS-BASED MOTION CORRECTION FOR ON-LINE MR TEMPERATURE MAPPING ............................. ,2571
Denis De Senneville Baudouin. /ME ERT CNRS, France; Quesson Bruno, Image Guided Therupy, France: Desbarars Pascal, LaBRI,
UMR 58W CNRS, France; Salomir Rnres, U386 INSERM, France; Pnlussiere Jean, lnstirur Bergonie, France: and C.T U! Moonen, IME
ERT CNRS. France
SHAPE PRIOR BASED SEGMENTATION FOR ORGAN DEFORMATION CORRECTION. . . . . . . . 2575
Jun Xie, H.T Tsui. Electrnnic Engineering Department, The Chinese University of Hong Kong, Hong Kong; and Lam Wai-Man, Faculty
of Medicine, The Chinese Universiry of Hong Kong, Hong Kong
TRACKING DEFORMABLE OBJECTS IN GEOSPATIAL APPLICATIONS ... . . . . . . ,2579
Sotirios Gvfakis, Peggy A p u & and Anthony Stefanidis, University of Maine, USA
A MOTION CONFIDENCE MEASURE FROM PHASE INFORMATION ............................................. .2583
I d n g To, Mark Pickering, Michael Frater and John Amold, Universiry College, The University of New Sourh Wufes,Australia
A NEW FEATURE CLUSTERING METHOD FOR OBJECT DETECTION WITH AN ACTIVE CAMERA, . . . . . . . . . . . . . . . . . ,2587
Christian Mic/ieloni, Gian Luca Foresri and Flavio Alberti, Deparrmenr of Compurer Science, Universiry of Udine, Iraly

xxxi
UNSUPERVISED MOTION DETECTION USING A MARKOVIAN TEMPORAL MODEL WITH GLOBAL
SPATIAL CONSTRAINTS. . . . . . . . . ................................
Pierre-Marc Jadoin and Max Mignotre, irv of Mmrrenl, Canada
DETECTION AND TRACKING OF MOVING OBJECTS IN IMAGE SEQUENCES . . . . . . . . . . . ,2595
Min Xu. Ruixin Niii nnrl Pruinod Varslrney, Dept. of EECS. Syracuse Univercirx USA
MOTION DETECTION BASED ON CONTOUR STRINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2599
Michael Kelliter and Tobias Hflnnblg. FORWISS, UniversiT of Passau, Germany
POSE ESTIMATION USING FEATURE CORRESPONDENCES AND DTM
Ronen Lemrc Eliud Ridin, The Teclmion, Depr. of Coirtprler Science. Israel:
Aerospuce Engineering and Mechanics, Minne.sura
MARKERLESS MOTION CAPTURE WITH SINGLE AND MULTIPLE CAMERAS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -2607
P0ri.s Kaiiiiokis and Joan Lnsenby Cambridge (ifliversin. Engineering Deparmenr, UK

SESSIONWA-P2: Watermarking V
ROTATION AND SCALE INSENSITIVE IMAGE WATERMARKING.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,261 I
Maxiiire Os.sonce, Claride Delplro and Pierre Dulrcimel, US, France
A PRACTICAL VERSION OF WONG'S WATERMARKING TECHNIQUE ..... ..... 2615
Abdelkader Oudo and Mrrh,wrd EI-Sdko, Unitanil? of Wrsrrnr o!flaI'i<J,Cam
IMPROVED BIT RATE CONTROL FOR COMPRESSED VIDEO WATERMARKING.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2619
Sugiri Pr"nrrrro. Vikror Wahaduniiilr, Yong Liang Guan and Hock Chuan Cliua. Na,iwng Tec/inological Univer.~ir~
Singapore
MODEL BASED STEGANALYSIS . . . . . . . . . ..... ...... . . . . ,2625
Xino)'i Re. Rohung Wang nitd Tienie Em.Narional Diboruron. of Parrrm Recognirion,lnsrirure of Aulonrarion, Chinese Acade,it? q j
Science. Cltino
AN IMAGE-QUALITY GUARANTEED METHOD FOR QUANTIZATION-BASED WATERMARKING USING A DWT.. . . . . . 2629
Masarrki Fujijoslri and Hiroslii Kiya, ToLy Metropolitan Universiry, Japan
A BLIND CDMA IMAGE WATERMARKING SCHEME IN WAVELET DOMAIN.. . . . . ....... . . . . ,2633
Sanri I? Mait?: B.E.College (D.U), Howrrrh, India; and MoIay K. Kundu, Indian Statistical Institere, India
AUTHENTICATION OF LOSSY COMPRESSED VIDEO DATA BY SEMI-FRAGILE WATERMARKING . . . . . . . . . . . . . . . . . . ,2637
Tsong-W Cken. Chien-Hua Hunang, Tkou-Ho Clten. Department of Elecrronic Enginewing, Narional Kaohsiung University of Applied
Sciences, Taiwan: and Clreng-Chieh Liu, Huper Laborarories Co., LTD, Taiwan
WATERMARKING FOR JPEG IMAGE AUTHENTICATION SURVIVING INTEGER ROUNDING IN DECOMPRESSION
Hiroshi /to, Koiclti Magai, Ryousuke Fujii and Mifsuyoski Suzuki, Mirsubislri Electric Corporation, Japan
A CLASSIFIER DESIGN FOR DETECTING IMAGE MANIPULATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2645
Istnail Avcibus, Sevinc Bayrum, Uludag Universiw. Turkey: Nasir Memo". Polyrecknic UniverriQ, USA: and Bulenr Sank",: Bogazicki
Universil): Turkey
A NOVEL DIGITAL IMAGE WATERMARKING SCHEME USING BLIND SOURCE SEPARATION. ..................... .2649
Thang I.! Nguyen, Jagdisk C. Palra and Norendra S. Clurudlurri, Nmyang Tecknological University,Singapore
A TRACEABLE CONTENTADAPTIVE FINGERPRINTING FOR MULTIMEDIA ............... 2653
Yu-Tzu Lin and Ja-Ling Wu, CML. Dept. ofCSIE, NTU,Taiwan
TRANSPARENT INFORMATION HIDING WITH AUTOMATIC EMBEDDING RANGE SELECTION FOR
OWNERSHIP VERIFICATION. . . . . . . . . . . . . 2651
Famok Sarrac Dan Yu, Sirajudeen Culam Razul
HIERARCHICAL MULTIPLE IMAGE WATERMARKING FOR IMAGE AUTHENTICATION AND
OWNERSHIPVERIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2661
Jagdisk C. Palra, Kah K. Ang and Ee-hang Ang, Nanyang Technologicai University, Singapore

SESSIONWA-P3: Feature Extraction and Analysis I

ESTIMATION OF MULTTPLE ORIENTATIONS IN MULTI-DIMENSIONAL SIGNALS. ................................ .2665
Cicrro Mora, llniversiry of Luebeck. Gemany; Til Aack. University of luebeck, Germany; lngo Stuke and Erkadt Earth, University of
Luebeck, Germany
CORNER DETECTION OF GRAY LEVEL IMAGES USING GABOR WAVELETS.. ................................... ,2669
xinring Cao, Farook Sarrac School of EEE, NTU, Singapore: and Ronda Venkaresworlu, Inrrirure for lnfocomm Research Singapore
A FAST M m H O D TO IMPROVE THE STABILITY OF INTEREST POINT D E E C l l O N UNDER ILLUMINATION CHANGES. . ,2673
Flore Faille. Tecknische Universitaet Muencken, Germany
THE ITERATlVE OBJECT SYMMETRY TRANSFORM.. ......................... ..,. 2677
Bertrand Znvidovique, IEF; University Paris XI, France; and vim Di Gesu', DMA, Universila'di Palermo, Italy
FEATURE SPACE ANALYSIS USING LOW-ORDER TENSOR VOTING. ............................................ .2681
Jia Wang, Hanqing Lu and Qingskan Liu. National Lnboratory of Parrern Recognirion, Institute ofAutomation, ChineseAcademy of
Sciences, Ckina

xxxii
EDGE AND LINE DETECTION AS EXERCISES IN HYPOTHESIS TESTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2685
Curry New.sarn, Defence Science & Tecltnologp Organisadon, Australia
SURFACE RADIANCE: EMPIRICAL DATA AGAINST MODEL PREDICTIONS 2689
Hossein Raghrb and Edwin Huncuck, Deparrment of Conrpurer Sciertce. Universiry of York, UK
OPTIMAL SEGMENTATION OF SIGNALS AND ITS APPLICATION TO IMAGE DENOlSlNG AND
BOUNDARY FEATURE EXTRACTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2691
Tony Han, University of Illinois, USA; Srevrn K q University of Rltode Island, USA; and TI~o,nnsHuang, University of Illinois, USA
KERNEL GENERALIZED NONLINEAR DISCRIMINANT ANALYSIS ALGORITHM FOR PATTERN RECOGNIT
Gunng Doi und Yunrao @an. College of Compurer Science, Zlzejiung Univerrirp, Chino
SOURCE SEPARATION IN NOISY ASTROPHYSICAL IMAGES MODELLED BY MARKOV RANDOM FIELDS
Ercan Kuruoglu. Anno Tonuz;ini nnrl Lnurri Bimcki, ISTI-CNR. lralv
WATER VIDEO ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2705
Lisrr Spencer and Mubrrrak Shah, Uniwrsity of Centrul Florida, 1J.U
FACE RECOGNITION USING REINFORCEMENT LEARNING 2709
Me/wrr,slr 7irfi,;;oli Hnrandi, Mnjid Nil; AImadab0di rrnd Babak N.
A PROBABILISTIC FRAMEWORK FOR OBJECT RECOGNITION IN VIDEO. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2713
Onmr Jawd. Muharok Sldr, Uniwrsity of Cenrml Flo,-ida, USA: m d Durin Cotnrmiciu, Siemens Corporare Resemdr, USA

SESSIONWA-P4: Image Segmentation: Level Set and Active Contour

SAR IMAGE SEGMENTATION WITH ACTIVE CONTOURS AND LEVEL SETS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2717
I s r n d nen Ayed. Carlos Wiryuei, Aiiur Milidie, INKS-EM[ Canada; orid Ziad Bdlradj, E w l e Supe,-ie~tre des Co,,airunicotions. T " i u
IMAGE PARTlONlNG BY LEVEL SET MULTIREGION COMPETITION.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2721
Abdol-Rcsn Mansoari, Amor Mirirlze nnd Carlos Vazyur;, INKS-EM? Canorla
INTERACTIVE SEGMENTATION USING CURVE EVOLUTION AND RELEVANCE FEEDBACK. ...................... ,2725
Motu; El Sabon and Banglore Manjrmullr, UCSB, USA
SHAPE GRADIENT FOR MULTI-MODAL IMAGE SEGMENTATION USING MUTUAL INFORMATION. . . . . . . . . . . . . . . . . ,2729
Ariune Her6rdo1, 13s Laborarory, France: Stipphunie Jellan-Besson, GREYC-imugesLaboraton,, Frunce; Michel Barlaud, 13s
kbomtory, France; and Gilles Aubert, Dieudonne Laboratory, France
MEDICAL IMAGE SEGMENTATION WITH MINIMAL PATH DEFORMABLE MODELS .............................. .2733
Pingkun Yon und AshraJKassim. National Universiry of Singapore, Singapore
AUTOMATED CELL NUCLEUS SEGMENTATION USING IMPROVED SNAKE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2737
Min Hu, Xijian Ping and Yihung Ding, lnfonnarion Engineering Universiry, China
SEGMENTATION OF ANATOMICAL STRUCTURES FROM 3D BRAIN MRI USING AUTOMATICALLYBUILT
STATISTICAL SHAPE MODELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2741
Jonurhan Bailleul, Su Ruan, Daniel Bloyet and Barbara Romaniuk, GREYC - CNRS UMR 6072, Frunce
CARTILAGE SURFACE TRACKING USING DIRECTIONAL GRADIENT VECTOR FLOW SNAKES .................... ,2745
Jinshan Tung. Depurrment of Elecrricul Engineering, University of Virginia, USA; Steven Millington, Center for Applied Biomechanics,
Deparrrnenr of Orthopaedic Surgery,Universiry of Virginia. USA; and Scott TAcron, Depurtment of Elecrricd Engineering, Universiry of
Virginia. U.TA
CONVERGENCE ANALYSIS OF ACTIVE CONTOURS IN IMAGE SEGMENTATTON ................................. ,2749
Rafael Verdu Monedero, Juan Morales Sanchez, Ricardo Gonmlez Leon, Universidad Polirimica de Cartageno, Spain; and Luis
Weruuga Priero, Ausrrian Academy oJSciences, Austria
REMESHING ALGORITHM FOR MULTIRESOLUTION PRIOR MODEL IN SEGMENTATION. .................... 2753
Gouaillurd Alexandre, Gelas Amaud. crearis, France; Vulette Sebastien, LIS, France; Bok Eric. cre ance; Kunai Takaslri, Keio
Universiry, Japan; and Prosr Remy, creatis, France
UNSUPERVISED SEGMENTATION ALGORITHM OF HRTEM IMAGES. ........................................... ,2757
Ainhou Mendizubal. Julian Cabreru, Luis Salgudo, Narciso Gamin and J w n Gonzalez, Universidad Politecnica de Madrid. Spain

SESSIONWA-PS: Transcoding
A COMPRESSED-DOMAIN HETEROGENEOUS VIDEO TRANSCODER. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,2761
Wan-Chi Siu, Kui-Tar Fung and Yui-Lam Chan, Hong Kong Polytechnic University, Hong Kong
AN IMPROVED RATE-QUANTIZATION MODEL FOR RATE CONTROL IN REAL-TIME VIDEO ENCODING
Bo Xie, PackdVideo Corporarim. USA: and Wenjun Zeng, Univ. of Missouri-Columbia, USA
UPIB FRAME TYPE DECISION BY COLLINEARITY OF DISPLACEMENTS
Adriuna Dumirras and Barry G. Huskell, Apple Compurez USA
RATE ESTIMATION FOR H.2WAVC SPATIAL RESOLUTION REDUCTION ........................................ -2773
Perer H. U! Wong, Roberr 7: M! Hung, Jack L B. Lee, S. C. Liew, C. S. Kim, The Chinese Universiry oJHong Kong, Hong Kong; and
Roland T Chin, The Hong Kong University of Science and Technology, Hong Kong

xxxiii
MPEG-21 DIGITAL ITEM ADAPTATION BY APPLYING PERCEIVEDMWION ENERGY TO H.264 VIDEO . . . . . . . . ,2777
Zltrio Gong, Departntent of Bcctronics & Informution ~ngbieeriiig,Hun:lrung University of Science & Technologx Chinu; l.inng-7ien
Cliiu, Centerfur Multimedia arid Network Technolog.v,Scliuol of Coinpuler Engineering, Nniivang Teclmulogicul University, Singapore:
and Yang Z x ~ g k u i Depnmrtznr
, of Electronics & Iitforinarion Engineering, Hunzhong Univerritv of Science & Technology, China
MODE MAPPING METHOD FOR H.264lAVC SPATIAL DOWN-SCALING TRANSCODINC. ........................... .2781
Peng Zlmng, Institute of Computing Technology, Cliinese Acoderny of Science, China; Yun Lu, Microsoft Keseorch Asia, Chino; Qingming
Hiwng und Wen Guo, Institute of Computing 7eclinoloyy, Chinese Acadeinv of Science, Cliinu
FASTTRANSCODING OF INTRA FRAMES BETWEEN H.263 AND H.264.. ....................... 278.5
Jens Uiulkowki, A n d r t Kmp, Uniwrsip of ~rlrmgm-Nuretiiberg,Germany; and Klaus Illgnec Siemens AG, Gensnn?
JOINT CONTROL FOR HYBRID TRANSCODING USING MULTIDIMENSIONAL RATE DISTORTION MODELING.. . . . . . ,2789
Yong Ju Jung and Yong Mrrn Ro, l n J ? m ~ i d o nand Consnrmicrrtions Univerril~:South Korea
FORMAT-INDEPENDENTSCALABLE BIT-STREAM ADAPTATION USING MPEG-21 DIA.. . . . . . . . . . .2793
Debarglza Muklrerjee, Gmrlrline Kuo, SId-to Hsinrrg, Sum Liu and Amir Said, Hewlett Packurd Irrhorriro
ON RESIZING IMAGES IN THE DCT DOMAIN.. . .. . . . . . . . . . . . . . . . . . . . ,2197
Curlos Salazar- and Trac 7rcm Tlie Johns Hupkins Universrtr:
RESIZING OF IMAGES I N THE DCT SPACE BY ARBITRARY FACTORS.. . . . . . . . . . . . . 2801
Joyonto Makhopodli,w~:lnrlion Institute of Teclinology, Kliomgpuc Inrlio: and Sanjir Mirrcr, Unh! of Crrlifornio, Suntu Uurbom. USA
GRAPHICS-TO-VIDEO ENCODING FOR 3G MOBILE GAME VIEWER MULTICAST USING DEPTH VALUES.. . . . . . . . . . ,2805
G e m Clzermg. Trrkaski Sakarr~oloand Wui-lion7kn. Hewlett-Packard Lrrlmrarories, USA
INTEGRATED COMPRESSED DOMAIN RESOLUTION CONVERSION WITH DE-INTERLACING FOR DV
TO MPEG-4 TKANSCODING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............. . . . . . . . . . . . . . 2809
H u r d i r a Knro. KODI R&D I~rburaroriesInc., Japan: Tiiknslri Sano. TA?o Universip of S<:ien<:e,Jupan; and Yaisqaki Nr,kijima, KDDl
K&D Lnbomrorirs Inc.. Japan

SESSIONWA-P6: Implementations and Systems

ARCHITECTURE OF MPEG-7 COLOR STRUCTURE DESCRIPTION GENERATOR FOR REALTIME
VIDEO APPLICATIONS . . .........
Jing- Ying Cliang, Grudurrre
Universit?: Taiwan; Liung-Gee Chen and Hung-Chi Fong, Graduote Institute of Electronics Engineering. Norionol Taiwan University,
Taiwan
IMPROVED THROUGHPUT ARITHMETIC CODER FOR JPEG .......................................... 2817
Michuel Dye< David Taubmon and Saeid Nooshabadi, University Wales, Ausrrnlia
DESIGN FLEXIBILITY USING FPGA DYNAMICAL RECONFIGURATION. . . . . . . ....... .2821
Nicolus Abel, Lounir Kessul m d Didier Dernigny, ETlS - UMR 8051 CNRS, France
DESIGN AND FPGA IMPLEMENTATION OF NON-SEPARABLE 2-D BIORTHOGONAL WAVELET TRANSFORMS
FOR IMAGENIDEO CODING ....... . . . . . 2825
Isu Seman Uzun and Abbes Ami 's Unive
AWOMATIC KEYSTONE CORRECTION FOR SMART PROJECTORS WITH EMBEDDED CAMERA . . . . . . . . . . . . . ,2829
Baoxin Li und M. Ibrahim Sezan, ShuT Labs of America, USA
HIGH-THROUGHPUTIMAGE ROTATIONUSING SIGN-PREDICTIONBASED REDUNDANT CORDIC ALGORITHM . . . . ,2833
Suchirru Surhyannruyana, Siew Kei Loin and Srikanrhun Tlzimbipillii, Nunyung Technologicul Universiry. Singapore
SMART IMAGE SENSOR FOR HIGH SPEED IN-FOCUS DETECTION. .... ................. .................. ,2837
Tukslri Yoshida, Arimiisu Yokota, Hideki Kashiyumu and Taknyuki Hamamoro, Tokyo Universiry of Science, Japan
IMPLEMENTATIONOF JPEGZMX)ARITHMETIC DECODER USING DYNAMIC RECONFIGURATION O F FPGA.. ....... ,2841
Sophie Boucliooux, Elbay Bourennane and Michel Puindavoine, LEX, Frunce
SCALABLE VIDEO CODING BASED ON MOTION-COMPENSATED TEMPORAL FILTERING:
COMPLEXITY AND FUNCTIONALITY ANALYSIS. ........ .... .2845
Fabio Verdicchio. Yiannis Andreopoulos, Tom Clerckr, Joeri Burbarien, Adrian Munreanu, Jan Comelir and Peter Sclrelkens, Vrije
Universiteit Brussel - ETRO, Belgium
A SMART CAMERA FOR FACE RECOGNITION , . .........
Richard Kleihorsr, Philips Researrh Labs, The Nethe
Harr). Broers, Pliilips C n : The Netherlands
OBJECT COLOR PROPAGATION IN AN UNREGISTERED DISTRIBUTED VIDEO SENSOR NETWORK . . . . . . . . . . . . . . . . ,2853
t4ju.y Venkararamun, Sabrshan Srinivusan und Anthony Srefunidis, Universiry of Maine, USA
ACCELERATlN,GTHE COMPUTATION OF GLCM AND HARALICK TEXTURE FEATURES ON
RECONFIGURABLE HARDWARE .............. ........................................... .2857
M u l r a m d Atif Tahi,: Ahmed Bouridane, Forih Kurugollu and Abbes Am Queens University Be@% U K
RATE AND DECODING POWER CONSTRAINED VIDEO CODING SCHEME FOR MOBILE MULTIMEDIA PLAYERS . . . . ,2861
Ligang Lu und Vudim Sheinin, IBM Wutson Research Cenlec USA

xxxiv
REALITY WINDOW MANAGER A USER INTERFACE FOR MEDIATED REALITY 2x65
Roscv Hill, James Fung and Slew Munn, University cf Toronto, Cunudrr

SESSIONWA-P7: Document Image Processing and Other Applications

EFFICIENT AND RELIABLE DYNAMIC QUALITY CONTROL FOR COMPRESSION OF COMPOUND
DOCUMENTIMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2x69
Ainir Said, Hewlett Prrckoml Labomrorirs. USA
CONTEXTUAL DISAMBIGUATION FOR MULTI-CLASS OBJECT DETECTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Xinodons Fm. Dept. of Ellecrricul and Cmnputer Engineering, The Johns Hopkins University, USA
DOCUMENT IMAGE RECTIFICATIONUSING FUZZY SETS A N D MORPHOLOGICALOPERATORS . . . . ,2877
Shijiun Lu, Uen M. Clien and C. C. KO, Nutionul Universirv of Singapore, Singapore
ON-LINE HANDWRIITEN CHINESE CHARACTER RECOGNITION USING A RADICAL-BASED AFFINE
TRANSFORMATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2XXl
Yuan-BiLii, Laii-Shing limo1 Yunlin Univ. of Science cind Teclr.. 2riwun
VISUAL ECHO CANCELLATION IN A PROJECTOR-CAMERA-WHITEBOARD SYSTEM.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288.5
Hniinin,q Zlioa, University of Illinois n r U~6",,~,-Cl,",,,,~',ig,~, (/.TA: Zh'I,EngwnZlrung. Micmsofl Rrsmn.1). USA; und Tliomrr,~Huang,
IJniverrity of Illinois ut Urba,ia-CBor,ipoign. U.S
ELLIPTIC A R C VECTORIZATION FOR 311 PIE CHART RECOGNITION . . . . . . . . . . . . . . . . . . . .2889
Weiliua Huong. Cliew Linz Em 'rrnd Wee Klrmg Lams SOC. Notional u,7iVerSil? o/Singu/,o,y, SinL
DOCUMENT IMAGE SECRET SHARING USING HIT-LEVEL PROCESSING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2893
Rristislrn, Lukuc "IKonsrnnrinos Plumnioris, Univer.~ir>,of E,,vrrro, Grnod~r
ADAPTIVE PHOTO COLLECTION PAGE LAYOUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2897
Clayton Arkin$ Hewlrrf-Puckord Lnbs, USA
A HIDDEN MARKOV MODEL FRAMEWORK FOR TRAFFIC EVENT DETECTION USING VIDEO FEATURES.. . . . . . . . . ,2901
Xiuokun 1.i. Universie of Cincinnati, USA: und Fufilr M. Porikli. Misubishi Elecrric Research Lobs, USA
AUTOMATIC TEXT SEGMENTATION FROM COMPLEX BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2905
Qiriung Ye, lnsrirure of Co,n,xiting Technology, Chino; Wen Guo and Qinyming Hunng, Grndriute School of Chinese Acudenry of
Sciences. China
TRACKING FOOTBALL PLAYERS WITH MULTIPLE CAMERAS .................... 2909
Ming Xu, Jumes Orwell and Craeme A. Jones, Kingston Universiry. UK
IMAGE CONTENTBASED ACTIVE SENSOR PLANNING FOR A MOBILE TRINOCULAR ACTIVE VISION SYSTEM . . . . .2913
Alan E. Abdel-Hakbn and Aly A. Furug, Universiry of Louisville, USA
IDENTIFICATION OF INSECT DAMAGED WHEAT KERNELS USING TRANSMIITANCE IMAGES. ................... ,2917
&lira Catalrepe. Siemens Corporate Researrl!, USA; Enis Cerin, Bilkenr University, Turkey; and Tom Pearson. US Depr OfAgriculrurr. USA

SESSIONWA-PS: Biomedical Image Processing: Segmentation and Computer Assisted Screening & Diagnosis
DETECTION AND RECOGNITION OF LUNG NODULES IN SPIRAL CT IMAGES USING DEFORMABLE
TEMPLATES AND BAYESIAN POST-CLASSIFICATION ............................................ 2921
Aly Farag, A p a n El-Bar. CVIP lab, Universiry of Louisville, USA; Gewrgy Cirnel'furb, University of Auckland, New Zeulund; and
Roberr Fulk. Jewish Hospiral, USA
3D MEDICAL IMAGE SEGMENTATION APPROACH BASED ON MULTI-LABEL FRONT PROPAGATION.. . 2925
Hwr Li. Grey-Ensicuen, F r n c e ; Abderrakim Elmoatoz. LUSAC. Sire Universiruire, France; Julal Fndili, Creyc- E m
Su Ruan. Equipe Image, L.A.M., France; and Bnrbura Romaniuk, Greyc-Ensicaen, France
CHROMOSOME COUNTING VIA DIGITAL IMAGE ANALYSIS ............................................. ,2929
Victor Cajendran and Jeffrey J. Rodriguez. Universiry of Arizona, U
QUANTITATIVE ANALYSIS OF LYMPHOCYTE MEMBRANE PROTEIN REDISTRIBUTION FROM
FLUORESCENCE MICROSCOPY. ............................................................................ ,2933
Perer Kusson, Johannes Huppa, Murk Davis and Axel Brungec Stanford Universiry School of Medicine, USA
CELL NUCLEI SEGMENTATION USING FUZZY LOGIC ENGINE. . . . . . . . . . . . . . . . . . . . 2937
Grigory Begelman, Computer Science Dept., Technion- lsruel Institute of Technology, Israel; E
University. Ismel: Ehud Rivlin, Michael Rudzshy, Computer Science Dept., Technion- lsrnel lnsrirute of Technology. Israel; and Zeev
Zalevsky, School of Engineering, Bar-Nan Universiry, Israel
TEXTURE CLASSIFICATION OF SARS INFECTED REGION IN RADIOGRAPHIC IMAGE. ........................... ,2941
Xiuoou Ring, Dacheng Tuo and Gregory E. Antonio, The Chinese University of Hong Kong, Hong Kong
BLOOD FLOW GENERATION IN B-MODE ULTRASOUND IMAGES OF THE CAROTID ARTERY ................ 2945
Ali Hamou and Muhmoud El-Sakkn, The Universiry of Western Ontario, Canada
ESTIMATION OF MIXTURES OF PROBABILISTIC PCA WITH STOCHASTIC EM FOR THE 3D BIPLANAR
RECONSTRUCTION OF SCOLIOTIC RIB CAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2949
Said Benameuc Ecole de technologie Superieure, Canada; Mar Mignotte, Franc,ois Destrcmpes, Universire de Monrreul, Canada; and
Jacques De Guise. &ole de rechnologie Superieure, Canada

xxxv
MICROCALCIFICATION DETECTION BASED ON LOCALIZED TEXTURE COMPARISON.. . . . . 2953
Xin YIian and Pmgckcng Ski, Hong Kony Universir). of Science and Tec/inology. Hong Kong
BREASTCANCER DIAGNOSIS USING IMAGE RETRIEVAL FOR DIFFERENT ULTRASONIC SYSTEMS.. . . . . . . . . . . . . . ,2957
Yu-Len Huang, Deporrmenr of Conipurer Science and bfonnarion Engineering Tunghai Uniwrsiry, Priwrin; Dar-Ren Clten, Depnrrmenr
of General Surge? Clrina Medico1 College & Hospiral, Taiwan: and Yu-Kuang Liu. Depurrmmt of Compurer Science and In/un,iorion
Engineering ncnylmi Universiq, Tniwan
GENDER DETERMINING METHOD USING THERMOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296 I
Satoslii Nislrino, Sachiyo lgaraslii ond A rsuslii Mal.suda, Oyuaro Nutionol Collegr of Trchnnlogy J a p n

SESSION WP-L1: Image Representation, Rendering, and Quality Assessment

CAMERA RESPONSE FUNCTION RECOVERY FROM DIFFERENT ILLUMINATIONS OF IDENTICAL SUBJECT MATTER . . ,2965
Corrv Manderr, Cliris Aiinonie ond MUIIIISleve, Universirv @ 7bronIo, Canadu
CLUSTERING-BASED MATCH PROPAGATION FOR IMAGE-BASED RENDERING FROM MULTIPLE IMAGES.. . . . . . . . .2969
Jian Y m und Wui-Kuen Cham, Dcpamnenr of Elecrmnic Engineering. The Chinese Uniwmiry of Hony Kong, Hong Kong
A TRAINING-BASED NO-REFERENCE IMAGE QUALITY ASSESSMENT ALGORITHM ............................. ,2973
Hiiirrro Luo, HP labs. USA
A CALIBRATED PINHOLE CAMERA MODEL FOR SINGLE VIEWPOINT OMNIDIRECTIONAL IMAGING SYSTEMS.. . . ,2977
Dunid Moldotwn and Tusltikozrr W~idu.W&~wr~r<tUniversir?, Jupun
AUTOMATIC DETECTION OF DIGITAL ZOOMS . . . . . . . . . . :. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 I
Jerome B u z i arid Frederic Guicliard, DO I*rbs, France
UNIQUENESS OF BLUR MEASURE ................................................. 29XS
Jerome Burzi and Fred+ Guichnrd, D
ARBITRARY VIEWPO[NTS IMAGE REPRESENTATION BASED ON 3-D GEOMETRIC MODEL AND
SPECULAR REFLECTION TRACING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2989
MasnrrM Endo, Tukalrim Marsrrslrira, Koichi Fukud~rand Akirrr Kuwrwonoku, Soplria Univerriw, Japan
STEREOSCOPIC IMAGE GENERATION BASED ON DEPTH IMAGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2993
Linng Zlrmg, Wu Jarnzs Tma and Demin Wang. Cornmenicarions Reseurclr Cmrre Canada, Canada

SESSION WP-LZ: Stereoscopic Image Processing and 3D Modeling

A LAYERED STEREO ALGORITHM USING IMAGE SEGMENTATIONAND GLOBAL VISIBILITY CONSTRAINTS ...... ,2997
Micfiael Bleyer und Margrir Gelautz, lnrerucrive Media Systems Group, Institute for Sofnvnre Technology and Interactive Sysrelns, Vienna
Universiry of techno log.^, Ausrria
AN ENERGY-BASED FRAMEWORK USING GLOBAL SPATIAL CONSTRAINTS FOR THE STEREO
CORRESPONDENCE PROBLEM ................................ ........................ 3001
Pierre-Marc Jodoin and Max Mignorre, University of Monrreal, Canada
STRUCTURED LIGHT SYSTEM CONFIGURATION DETERMINATION FOR EFFICIENT 3~ SURFACE RECONSTRUCTION.. ,3005
Alberr Dipanda and Sangli.vuk Woo, Laboraroire LE2I. France
VIRTUAL VIEW SYNTHESIS THROUGH LINEAR PROCESSING WITHOUT GEOMETRY ............................. 3009
Akira Kubora, Kiyoharu Aizawu, Univ. of Tobo, Japun: and Tsulnn Chen, Camegie Mellon Univ., USA
OPTIMAL HIERARCHICAL REPRESENTATIONAND SIMULATION OF CLOTH AND DEFORMABLE OBJECTS
Konsranrinos MOUSIR~US, Arisrotle University of Thessuloniki, Electrical und Computer Engineering Dept., Greece; Dimirrios Tzovarus.
Informatics and TelemuficsInsrirufe, Greece; and Michael G. Srrintzis, Arisrorle University of Thessaloniki, Electrical and Computer
Engineering Dcpr., Greece
ROBUST RECONSTRUCTION OF 3D POINTS FROM IMAGES ............................. 3017
Rui Rodrigues and Ant bnio Femandes, Universidade do Minho, Par ugu
3-D GEOMETRY ENHANCEMENT BY CONTOUR OPTIMIZATION IN TURNTABLE SEQUENCES .................... ,3021
Perer Eiserr, Fruunhofer Insrirure for Telecommunicarions,Heinrich-Herrz-lnsrirure, Germany
IMPROVEMENT OF PHASE-BASED ALGORITHMS FOR DISPARITY ESTIMATION BY MEANS OF
MAGNITUDE INFORMATION .................................. 3025
Udo Ahlvers and Udo Zoelze,: He ed Forces, Depamnenr of Signal Pmcessing
and Cormnunicorions, Germany

SESSION WP-L3 Feature Extraction and Analysis I1

COLOR INVARIANT DENSITY ESTIMATION FOR IMAGE SEGMENTATION AND OBJECT TRACKING 3029
The0 Gevers and Frank AldershoE University of Amsterdam, The Nerherlnnds
COMPACT ROTATION-INVARIANT TEXTURE CLASSIFICATION ................................................ .3033
Paul Southam and Richard Hu~~vey,University of East Anglia, U K
OPTIMAL MULTIRESOLUTION WLYGONAL APPROXIMATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3037
Alexander Kolesnikov and Pas; Franli, University of Joensua, Finlund

xxxvi
AFFINE-INVARIANT CURVE MATCHING ............................................................... .3041
Mom0 Zuliani, Sitaram BIta,pvatlty, Deparrm Iecrrical and Computer Engineering, UniverciQ of Califumin, Sanra Borham, USA;
B. S. Mnnjunarlr, Deporrtrtrnl of Electrical and Cornpurer Engineering, Universir).of Cal$omia, Scinro Barham, USA; and C. S. KenneJ’,
Deparlinent of Elrclrical and Computer Engineering, UniversiQ of California, Sanra Burliam, USA
MOVING SHADOW REPRESENTATION BASED ON A LEVEL CURVES DECOMPOSITION . . . . . ,3045
Henri Nicolns, INRIA, France
A NOVEL IMAGE RECOGNITION METHOD BASED ON FEATURE-EXTRACTION VECTOR SCHEME . . . . . . . . . . . . . . . . ,3049
Masao Hirrenoro, Panasonic I,~u-riration SWICI~ISCo.,Lid, Japan; Takchiro O ~ O WandC L Miki H a s e y ” , Hokkaido Univerriy, Japon
FEATURE POINT EXTRACTION USING SCALE-SPACE REPRESENTATION ........................................ 3053
Yousri Abdrljrroued and Touradj Ebrdrimi, Swiss Fedem1 Institute of Tecltnology, S w i t z e r l w ~ I
QUATERNION WAVELETS FOR IMAGE ANALYSIS AND PROCESSING.. . . . . . . . ............. 3057
Wui L o n Clran. Hveoklio Clroi ond Riclmrd Bnmniuk, Rice Uniaersiry, USA

SESSION WP-L4: ImageNideo Segmentation and Tracking

BACKGROUND MODELING AND SUBTRACTION BY CODEBOOK CONSTRUCTION ................... 306 I
i i i m Co,npnrrr Vision Lob, Uniwrsiry <$
K ~ i i g i i ~ i ,K Mai$und, USA: Tlronnrar Clioliil~~hhong.sc, Faculty of Infi,nnotioir ?i?clinoIogy,
King Morrgkur:~Imtitiar of Teclinology, Tliailaml; David Ha,woorl and Lrm? D m i ~ Conipurw , Vision I d 7 , University of Manlrmd, USA
A DETECTION-BASED MULTIPLE OBJECT TRACKING METHOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3065
Mei Hnu, NEC Lnhs America, USA; Amit Sethi, UIUC, USA: Wci Hua and Yilrong Gong, NEC Iahs Anrerim, USA
RECOVERING FIELD OF VIEW LINES BY USING PROJECTIVE INVARIANTS.. .
Senein Kdipmalur-and Wayne Wolf: Princeton Unitwrsity, USA
OITIMAL SENSOR SELECTION FOR VIDEO-BASED TARGETTRACKING IN A WIRELESS SENSOR NETWORK . . . . . . ,3073
Pe’Eshalu Pukdawatra, Tlwasp~oulr,.~ P a p p s and Aggrlos K a t . ~ ~ i ~ g e INortitwestem
~is, Unioersir)., USA
MORPHOLOGICAL SEGMENTATION PRODUCES A VORONOI TESSELATION OF THE MARKERS.. . . . . . . . . . . . ,3077
Fernand Mnyer, Ecole des Mines de Paris, Fvnncr
GRAPH-BASED REPRESENTATION FOR 2-D SHAPE USING DECOMPOSITION SCHEME.. ......................... ,3081
Duck Hoon Kim, Seoul Narionnl Univer.sir). Republic of Korea; I1 Dong Em, Hankak Universiry of Foreign Studies Republic of Korea;
and Song Uk h e , Seoul Narional Universiry, Republic of Korea
PATH-BASED MORPHOLOGICAL OPENINGS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3085
Henk Hcijmrms, CWI. The Netherlands; Michael Buckley and Hugues Talbor, CSIRO, Ausrralia
MORPHOLOGICAL GRADIENT OPERATORS FOR COLOUR IMAGES ....................................... 3089
Adrian Evvnns, Universiry of Bath, UK

SESSION WP-L5: Distributed Source Coding and Scalability

VIDEO MULTICAST OVER LOSSY CHANNELS BASED ON DISTRIBUTED SOURCE CODING. ..................
Abhik Majumdur and Kannan Romclmndran, U.C. Berkeley, USA
WYNER-ZIV VIDEO CODING WITH HASH-BASED MOTION COMPENSATION AT THE RECEIVER
Anne Aaron. Shantanu R a m ond Bernd Cimd. Stanford Universiry, USA
SYSTEMATIC LOSSY FORWARD ERROR PROTECTION FOR ERROR-RESILIENT DIGITAL VIDEO
BROADCASTING - A WYNER-ZIV CODING APPROACH ................................... ..3101
Shanranu Rune, Anne Aaron and Bemd Girod, Stanford Universi
DISTRIBUTED CODING OF MULTISPECTRAL IMAGES: A SETTHEORETICAPPROACH . . . . . ,3105
Xin Li, LDCSEE. WVU, USA
ANALYSIS OFTHE EFFICIENCY OF SNR-SCALABLE STRATEGIES FOR MOTION COMPENSATED VIDEO CODERS . . ,3109
Josep Pmdes-Nebor, Departamento de Comunicociones. Universidad Polire!cnica de Valencia, Spain; Greg07 Cook, School of Medicine.
Indiana Universiry. USA; and Edward Delp, Video and Image Processing Laboratory. Purdue Universir), USA
SNR-SCALABLE EXTENSION OF H.264IAVC.. .................................................. ..... ,3113
Heiko Schwarz, Derlev Marpe and Thomas Wiegand, Fraunhofer lnsriturefor Telecommunications - Heinrich H e m Instirute, Germany
UNIFORM MOTION-COMPENSATED513 FILTERBANK FOR SUBBAND VIDEO CODING ................... .3117
Cr’egoire Pau and B’eaeatrice Pesquer-Popescu, ENSZ Frunce
RATE-DISTORTION BOUNDS FOR MOTION COMPENSATED RATE SCALABLE VIDEO CODERS. .................... -3121
Gregoty W Cook, Purdue University, USA; Josep Prades-Nebot, Universidad Polircnicia de Valencia, Spain; and Edward J. Delp.
Purdue Univeristy, USA

SESSION WP-L6 Video Streaming

A HYBRID WAVELFT FRAMEWORK FOR MODELING VBR VIDEO TRAFFIC ..................................... ,3125
Min Dui, Dmirri b g U i I I O V , Texas A%M Universiry. USA; and Hayder Radha, Michigan Slate Universiq, USA

xxxvii
MULTIPLE DESCRIITION CODING WITH ERROR CORRECTION CAPABILITIES: A N APPLICATION TO
MOTION JPEG 20(M.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Tammam 7ill0, M<r,-cu Grangerto and Gabriella Olmu. Politecnico di Torino, IIuly
OVERLAY MULTI-HOP FEC SCHEME FOR VIDEO STREAMING OVER PEER-TO-PEER NETWORKS . . . . . . . . ,3133
fitfeng Slmn. Kensselrrer Pol?technic lnsrirure. U5A; Ivan Bajic. Universip of Miami, USA: Shivkumor Kalyan
Renssclaer Polytechnic In.vtitute, USA
EFFICIENT PATH AGGREGATION AND ERROR CONTROL FOR VIDEO STREAMING . . . . . . . . . . .
Omesh lickoo, Sliivkunior Kalyanarci,non and Julm WO&, Rrnsrrber Polytechnic Institute, 1JSA
RATE-DISTORTION OPTIMIZED STREAMING FOR 3-D WAVELET VIDEO .....................
Cliuu-Ling Cltang, Sangeun H m and Bernd G i r d , Stanford University, USA
RATE-DISTORTION OPTIMIZED STREAMING OF VIDEO WITH MULTIPLE INDEPENDENT ENCODINGS . . . . . . . . . . . . ,3145
Mark Kulmnn and Benirl Cirod, Strmford Univer.siry, USA
JOINT SERVEWPEER RECEIVER-DRIVEN RATE-DISTORTION OPTIMIZED VIDEO STREAMING
USINCASYNCHRONOUS CLOCKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3l4Y
Drrnjiie Li, Dep/,artmentof ECE, Univrrsily of California, Davis, USA; Gene Clreung, HP 1uborrrtorie.v.Jupon. Japan; Clien-Nee Clruali
m d S.J.Ben RIO, Departnimr of ECE. Uniimsify of Califbrnirt, Dovi.7. USA
MDC AND PATH DIVERSITY IN VIDEO STREAMING.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3153
Siva . ~ , ~ , t ~ f ~ . s ~ ~K.
~P~ ~S~~bbolaksl~irti,
l ~ i ~ - ~ , ~ ~ ~ , Dqi~irti~ienr
of Electrical and Comprrter Engineerbrg,St ns Insrirrrre of 7i.chnulogy.. USA: m d
R.N. Umu, Deportment @'Co,nputer Sc:ienre, University of Texas cif Dalkrs,, USA

SESSION WP-PI: Image Coding I1

IMAGE CODING WITH ITERATED CONTOURLET A N D WAVELET TRANSFORMS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3157
Vivien Clzrrppelio: IRISA/Univer.vit 'e de Rmnes 1. France; Cliristine Guillemot m d Slovicn Mariitkovic, IRISA, Frmcr
FAST VQ ENCODING ALGORITHMS USING ANGULAR CONSTRAINT . . . . . . . . . . . . . . . ............... 3161
Kousrlr I m m m m , Ahmed Swilem and Hide" Haslrimoro, Kanazawa University, J o p m
WAVELETIMAGECODING BYDILATION-RUN ALGORITHM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3165
Zheng Wu and Mingyi He, Norrlrwesrern Polytechnical Universiry, China
AN IMPROVED EZW ALGORITHM BASED ON SET PARTITIONING I N HIERARCHICALTREES USING
WAVELET REGULARIT ......... ......... . . . . 3169
Sergio Penedu and Rui Se , Brazil
CONCURRENT ENCODING IN HIERACHICAL TREES FOR WAVELET BASED IMAGE COMPRESSION . . . . . . . . . . . . . . . . 3 173
ling-Xin Wang, Dept. of CSIE. National Cheng-Kung Universiry. Tainan, 7aiwan. Taiwan; Fang-Hsuan Clieng, Dept. of CSIE. Chung-
Hwa Universiry, HsinChu, Taiwan, Tuiwan; and Alvin W ) : Su, Dept. of CSIE. National Cheng-KUng UniversitJ: Tainan, Taiwan, Taiwan
IMPROVED FAST SEARCH METHOD FOR VECTOR QUANTIZATION USING DISCRETE WALSH TRANSFORM . . . . . . . -3177
Zltihin Pan, Koji Kotani and Taduhim Ohmi, Tohoku University. Japan
PRACTICAL LOW BIT RATE PREDICTIVE IMAGE CODER USING MULTI-RATE PROCESSING AND
ADAPTIVEENTROPYCODING ............................................................................... 3181
Anna N. Kim and Tor A. Ramstad, Norwegian Universiry of Science and Technology, Nonvuy
Cl' IMAGE COMPRESSION WITH LEVEL OF INTEREST.. ...... 3185
Masayuki Hashimoro, Kenji Matsuo. Atsuski Koike, KDDl R&D :Hirok
Information and Communications Technology. Japan: and Tetsuo Shimono, Hokkaido Tokai Universify, Japan
WAVELET-BASED CONTOURLET TRANSFORM AND ITS APPLICATION TO IMAGE CODING. ...................... .3189
Rumin Eslumi and Hayder Rudha, Michigan State Univers;@, USA
L1 NORM RASED SECOND GENERATION IMAGE CODING . . . . .3193
Morco Dalui and Riccardo Leonardi, Department of Electronics
AN EFFECTIVE FRACTAL IMAGE CODING METHOD WITHOUT SEARCH. ............................. 3197
Shen Furun and Osamu Hasegawa, T o b o Institute of Technology, Japan
MULTIDIMENSIONAL SIGNAL COMPRESSION USING MULTI-SCALE RECURRENT PAlTERNS WITH
SMOOTH SIDE-MATCH CRITERION.. ........ ~............ ....................
Eddie Filho, Genius Institute of Technology, Brazil; Murilo De Carvalho, TECKTC Universidode Federal Fluminense. Brazil; and
Eduardo Da Silva, Progruma de Engenharia EIPricdCOPPE Universidade Federal do Rio de Janeiro, Brazil
COLOR AND R.O.I. WITH JPEG2000 FOR WIRELESS VIDEOSURVEILLANCE. ..................................... ,3205
Franck Lurhon and Brice Beaumesnil, LIUPPA Laboratory, Frunce

SESSION WP-PZ: Source-channel Coding II

OPTIMIZATION OF H264 FOR LOW DELAY VIDEO COMMUNICATIONS OVER LOSSY CHANNELS ................. , 3 2 0 9
Oztun Hurmunci and Murat Tekalp, Universiry of Rochester; USA
POLYPHASE SPATIAL SUBSAMPLING MULTIPLE DESCRIPTION CODING OF VIDEO STREAMS WITH H 2 6 4 . . . . . . . . . .3213
Riccardo Bemrdini, Mar0 Durigon, Robeno R i d d o , Universily of Udine, Italy; Luca Celerro and Andrea Wtuli, STMicroelectmnics. Italy

xxxviii
SELECTIVE FEC FOR ERROR-RESILIENT IMAGE CODING ANDTRANSMISSION USING SIMILARITY
CHECK FUNCTIONS ......................................................... 3217
Tuyer-Tmng Lroil. Lina J. Karum, Rid<,A. Brrzzi. Ariwna Stare Uniwrsiv, USA; and Glen P Abouslemon. General Dynamics C4 Swtems,
USA
ITERATIVE JOINT SOURCE-CHANNEL DECODING OF VARIABLE LENGTH ENCODED VIDEO SEQUENCES
EXPLOITING SOURCESEMANTICS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3221
Hang Npyen, Alcrrtrl R&l. Fruncr; and Pierre Dukrirnel, CNRSNSS-Supelec. France
A NOVEL VISUAL DISTORTION SENSITIVITY ANALYSIS FOR VIDEO ENCODER BIT ALLOCATION . . . . 322s
Chili- Wei 7iing. l l e ~ ~ a r t ~ n eofn Eluctronics
t Engineering. National Chiao Tung Univuxity, Tniwan; Clting-Ho Clren. Kt-Hui Kt and Clrun-
Jen Tsni. Deportment of Coinpurer Sciencc and I,l/oiamtion Engineering, National Chiao Tung University, Tniwon
A NEW MULTIPLE DESCRIPTION LAYERED CODING METHOD OVER AD-HOC NETWORK.. ...................... ,3229
ScJng Xino, ISN Nutionrrl Ku? Lnb., Xidim fJniwr.iit?, Clrina: C1wngk.r Wu, ISN Nnrionol kej Lib.. Xidian University Clrina; and Fang
Zlinng, ISN Notional Kev lab., Xidian Universin: China
A LOW-COMPLEXITY SOFT VLC DECODER USING PERFORMANCE MODELING. ................................ ,3233
Tsu-Ming Lilt and Clim-Yi Lee, In.stirure ofE1ectmnic.i Engineering, National Chiao Tung UniwrsitN %wan
SLICE GROUP BASED MULTIPLE DESCRIPTION VIDEO CODING USING MOTION VECTOR ESTIMATION 3237
Dong Wang, Nislran Canagorajah and David Bull, University of Bristol, UK
A PARA-PSEUDO INVERSE BASED METHOD FOR RECONSTRUaION OF FILTER RANK FRAME-EXPANDED
SIGNALS FROM ERASURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3241
Ruvi Motw.iini, DSI. Singopow; and Clrrirrbie Guillemot, IRISA. INRIA, France
OPTIMIZED FILTERING AND RECONSTRUaION IN PREDICTIVE QUANTIZATION WITH LOSSES . . . . . . . . . . . . . . . . . ,3245
Alsson /;ltTclre,: fJniver.sip of Calijnmio, Berkele?, USA; Sundeep Rrmgon, Flovioir Techriologies. USA: Viwk Goyal. M17: USA; and
Konnon Raidrandrnn, Universin. of Calijomia, Berkelej, USA
A JOINT SOURCE-CHANNEL DISTORTION MODEL FOR JPEG COMPRESSED IMAGES ............................ ,3249
Mahnnoimad .%hi,: Hatiid Sheikh, Robert Heath and Alan Bovik, The Universiry of Texas at Austin, USA
CHANNEL MODELING AND ITS E F F E m ON THE END-TO-END DISTORTION IN WIRELESS VIDEO COMMUNICATIONS . , 3 2 5 3
Eren Soyak, Ingsnient Technologies, USA; Yiifrnch Eirenberg, Fan Zlrai. Randall Bern: Tlrmsyvorrlus Pappas and Aggelos Kutsaggelos,
Nortliwesrrm Universiry. USA
OFTIMAL OBJECT-BASED VIDEO COMMUNICATIONS OVER DIFFERENTIATED SERVICES NETWORKS.. . . . . . . . . . . .32S7
Haolrong Wang, Fan Zhai, Yiftacll Eisenberg and Aggelos Katsaggelos, Nortltwestem University, USA

SESSION WP-P3: Stereoscopic and 3-D Coding

UNCONSTRAINED FREE-VIEWPOINT VIDEO CODING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Edouard Lamboray, Michael Waschhusch, Stephan Wumrlin, Murkus Gross, ETH Zurich, Switzerland; and Hansperer
Research Lab, USA
AN EFFICIENT CODING FOR 3-D GEOMETRY DATA BASED ON SURFACE SIMPLIFICATION AND
WAVELET TRANSFORM ............................... 3265
Himki Amada, Kenji Kasai. Yutaka Saito, Kouichi Fukuda and Ak
SHARING OF MOTION VECTORS IN 3D VIDEO CODING.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3271
Stefan Grrwrrtsch and Erika MLller: University of Rostock. ln.stiture of Communications Engineering, Gennany
SEGMENTATION BASED DISPARITY ESTIMATION USING COLOR AND DEPTH INFORMATION ..................... .3275
Sang Yoon Park, Sang Hwa Lee and Nom I k Cho, Seoul National UniversiQ, South Korea
TECHNIQUES FOR IMPROVING STEREO DEPTH MAPS OF FACES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3279
Jason Baker: Vinod Chandran and Sridlm Sridharan, Queensland Universiry of Technology, Australia
LOSSY-TO-LOSSLESS BLOCK-BASED COMPRESSION OF HYPERSPECTRAL VOLUMETRIC DATA. . . . 3283
Xiaoli Tang and William Peorlmn, Rensselaer Polytechnic Institute, USA
FREE VIEWPOINT VIDEO EXTRACTION, REPRESENTATION. CODING, AND RENDERING ......................... ,3287
Aljoscha Smolic, Karsten Mueller: Philipp Merkle. Tobias Rein, Peter Eiserr, Thomas Wiegand and Matthias Kautmer: Fraunhofer HHI,
Germany

SESSION WP-P4: Super-resolution and Mosaic

RESTORATION AND DEMOSAICING FOR PIXEL MIXTURE IMAGES IN DSC VIDEO CLIPS ..................... .3291
lkuko Tsubaki and Kiyoharu Aizawa, University of T o b o , Japan
SINGLE-FRAME TEXT SUPER-RESOLUTION: A BAYESIAN APPROACH. ......................................... .3295
Gerald Dalley, Mitsubishi Electric Research Labs, USA; Bill Freeman, Massachusetts Institute of Technology, USA; and Joe Marks,
Mitsubishi Electric Research Labs, USA
NON RIGID REGISTRATION OF SHAPES VIA DIFFEOMORPHIC POINTMATCHING AND CLUSTERING.. . . . . . . .
Laurent Carcin. lnstitut Geographique National, France: Anand Rangorajan, University of Florida, USA; and Laurent Younes. Johns
Hopkins Universily, USA

xxxix
ENHANCED MOSAIC BLENDING USING INTRINSIC CAMERA PARAMETERS FROM A ROTATING
AND ZOOMING CAMERA . . . . . ............ ............ ...........
Dae-Woong Kim and Ki-Song Hong, orea
FACE ALIGNMENT USING INTRINSIC INFORMATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3307
Yuchi Huang, N r r r i o ~Lob
l of Parrem Rccognirion, China; Sleplren Lin. Microsofr Researclr Asia, Cltino; Hmying Lu. Nnrional Lnb of
Parrem Rrcognirion. Chino; and Heung- Yeung S l ~ t ~ m ,
Microsoft Research Asia China
WAVELET-BASED COLOR FILTER ARRAY DEMOSAICKING ..............
Jef Driesen und Paul Sclreundrrs, Universiry of Antwerp, Belgium

SESSION WP-P5: Image Formation I1

A 2-LEVEL DOMAIN DECOMPOSITION ALGORITHM FOR INVERSE DIFFUSE OPTICALTOMOGRAPHY . . . . . . . . . . . . ,3315
I / - Young Son. Mum Gaven. Birsen Yaizici, Renssefaer Pul~rcclinicInsfirure, USA; and Xovier Inter, Univrrsity ofPennsjlvr,nia, 1JSA
IMAGE FUSION USING WEIGHTED MULTISCALE FUNDAMENTAL FORM ..... . . ,3319
Tao C l r m RaoSan Ciio and Silong P m g , lnsrirme of Auromotion, C1,ine.w Acaden
SPECKLE IMAGE ANALYSIS OF CORTICAL BLOOD FLOW AND PERFUSION USING TEMPORALLY
DERIVEDCONTRASTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3323
Allon Tan. Wunzlren Lirr, Yan Smg Elijali Yew, Joseph Siire.dr Paul ond Sim Heug Ong. Depumnrnf of Electrical nnd Cmnprrer
of Biomgineering, National Uriivrrsitv of Singapore, Sirigriporc,
Enginee,-ing/Dii~isi~,rr
HYPERSPECTRAL TARGET DETECTION USING KERNEL MATCHED SUBSPACE DETECTOR . . . .3327
Heesung Kwoii and Nasscr M. Nasrabadi, US Ann? Resenrclr Lnboraton, USA
HYPERSPECTRAL ANOMALY DETECTION USING KERNEL RX-ALGORITHM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3311
Heesurig Kwon and Nusser M. Nosmbodi, US A n t ~ vResearch Lnbomron, USA
A NEW APPLICATION OF TEXTURE UNITCODING TO MASS CLASSIFICATION FOR MAMMOGRAMS.. . 333s
k m Chm and Chein-IClwng. UMBC, USA
DISCRIMINATION AND IDENTIFICATION FOR SUBPIXEL TARGETS IN HYPERSPECTRAL IMAGERY. . . . . . . . . . . . . . . .3339
Chein-I Cliang. Wrbnin Lilt and Clrein-Chi Clung, UMBC, USA
A GENERIC METHOD FOR GENERATING MULTISPECTRAL FILTER ARRAYS ... 3343
Lirlan Miao, Hairong Qi, University ofTennessee, USA; and We.sley E. Snyder; North a Sl
A COMPARATIVE STUDY OF STATISTICAL AND NEURAL METHODS FOR REMOTE-SENSING IMAGE
CLASSIFICATION AND DECISION FUSION ............ .......... ........... ........... ,3347
Safan Malrmoud, National Aurhority of Remote Sensing and Space Science, Egypt; Moumen El-Meleg", Assiur University, Egypf; and Aly
Farag, UniuersiQ of Louisville. USA
COMPARING COLOR AND TEXTURAL INFO LUTION SATELLITE
IMAGE CLASSIFICATION. . . . . . . . . . . . . . . . . ...........
Ewour Vansleenkisre,Abram Schourreer, Sidharra
MULTISENSOR RASTER AND VECTOR DATA FUSION BASED ON UNCERTAINTY
Sang-Chul Lee and Peler Bajcsv, N a r i o ~Center
l for Supercomputing Applications,

SESSION WP-P6: Motion Detection and Estimation: Other Methods

THE EFFECT OF GLOBAL MOTION PARAMETER ACCURACIES ON THE EFFICIENCY O F VIDEO CODING . . . . . . . . . . ,3359
Gokce Dane and Tmong Nguyen. University of California, San Diego, USA
EXTENSION OF AAM WITH 3D SHAPE MODEL FOR FACIAL SHAPE TRACKING .... ...... 3363
Jaewon Sung and Daijin Kim, POSTECH, South Korea
A ROBUST AND NON-ITERATIVE ESTIMATION METHOD OF MULTIPLE 2D MOTIONS ............................ ,3367
Eun-Young Kang, Califomia Slate Universiy, Los Angeles, USA; lsanc Cohen and Gerard Medioni, University of Sourliem California USA
GRADIENT BASED DOMINANT MOTION ESTIMATION WITH INTEGRAL PROJECTIONS FOR REAL TIME
VIDEO STABlLlSATlON . . . . . . . . . . . . . . . . . . . ........... 3371
Andrew Crawford, Hugh Denman, Francis Kelly, Francois Pitie and Ani1 Kokaram, Trinity Col
ROBUST MOTION-BASED IMAGE SEGMENTATION USING FUSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3375
Miclue1 Farmer; Eaton Corporarion, USA; Xiaoguang Lu, Hong Chen and Ani1 Jain, Michigan Stare University, USA
ON AUTOMATIC DETERMINATION OF VARYING FOCAL LENGTHS USING SEMIDEFINITE PROGRAMMING 3379
Motilal Agrawal, SRI Inremotional. USA
A SIMPLIFIED METHOD OF ENDOSCOPIC IMAGE DISTORTION CORRECTION BASED ON GREY
LEVEL REGISTRATION. ......... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3383
Roseber Miranda-Luna. Walrer Blonde Daul. Yahir Hemande:, Ruben Posada and Didier WOK CRANINPL France
GEOMETRIC CONSTRUCIlON OF THE CAUSTIC CURVES FOR CATADIOPTRIC SENSORS . . . . . . . . . . . .3387
Siohoi leng, Laborarory of Insrruinenfs and Sysrems of Ne de France (LISIF)/Luborarory of Complex Sysrems Conrrol, Analysis and
Communication (LACCSC) - ECE, France; and Ryad Benosman, Laboratory of lnsrrumenrs and Sysrems of /le de France. France

XI
FOCAL LENGTH SELF-CALIBRATION BASED ON DEGENERATED KRUPPA‘S EQUATIONS: METHOD
AND EVALUATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3391
Weilrrn Lao. lrisriturcfor Infocomm Reseurch, Singopore: Zlroolin Clteng, Depomnent of Meckrmicol Engineering, Nurionul University of
Singapore, Singcipore: Alvin Kurn, Institute for Infocomm Rescorch, Singapore: Tele Trn, Deportment of Computing, Currin Univercity of
Teclinolog~:Austrulia: and AsIira/ Kassini, Departrirent of Elecrricul rind Computer Engineering, National Universitv of Singapore.
Singapore
LINEAR CAMERA AUTOCALIBRATION WITH VARYING PARAMETERS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3395
Antonio Vuld4s, Drpur.tu,nmto de Geometrirr y Topologia. Universidad Corn11lutens~ de Mulrid, Spain: Jose 1. Rondu nnd Cuillerriio
Gallego, Deportaraento dc Senules, Sirrrrnos y Rndiioco,rirmicrrciunw. Universidrrd PolirGcnicrr de Modrid, Spain
MEAN-SHIV BACKGROUND IMAGE MODELLING 99
moss in^^ Piccrwdi and Ton? Jan. University of Technolog
FUSING VIDEO AND SPARSE DEPTH DATA IN STRUCTURE FROM MOTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3401
Qilvng U i o n g m d Robert Pless, Coszpater Science D q X . W<i,rhbIglonUnivercil?, USA
MOTION BLUR REMOVAL AND ITS APPLICATION TO VEHICLE SPEED DETECTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3407
Huei-Yung Lin <ind Kun-Jlzih Li, Ndionol Clrung Climg Universily, Ilepurtinenr of Elrctricrd EnginrerinR, T f l i ~ ~ i i
GLOBAL MOTION ESTIMATION FOR MPEG-ENCODED STREAMS . . . . . . . . . . . 3411
Renan Corrdr~ivmil Bernord Besserec University Lrr Rocldle, France

SESSION WP-W: Watermarking and Cryptography

FILTER BANK SELECTION FOR THE OWNERSHIP VERIFICATION OF WAVELET BASED DIGITAL
IMAGE WATERMARKING. . . . . . . . . . . . . . . .3415
Min-Jeri Tsni, Notional Cltino Tung Universit?: Tr,iwan
REVERSIBLE DATA-EMBEDDING WITH A HIERARCHICAL STRUCTURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3419
Jun Tim nnd Rqwtond Wells, lnremorionnl Urrivcrsih. Bwmen, Genriun?
WATERMARKING ROBUST AGAINST ANALOG VCR RECORDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3423
Koiclii Mogni, Hiroslii /to, Mitsuyoshi Nrzuki, Kohtum Asrri, Mitsubislri Electric Corporation I n ~ ~ n ~ ~Technology
~ t i o ~ t R&D Cenrec
Jupon; and Hidcruslri Mi.shinia, Mitsubishi Electric Corporation Advanced techno log^ R&D Centec Jupm
MULTIRESOLUTION FRAGILE WATERMARKING USING COMPLEX CHIRP SIGNAL FOR CONTENT
INTEGRITY VERIFICATION ...................... 3427
Dun Yu, Farook Sattar and Braham Barkat, Nanyung Technological University, Singapore
ENCRYPTION OF WAVELET-CODED IMAGERY USING RANDOM PERMUTATIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,343 I
Roland Norcen and Andreas UId, Sulrburg Universitv. Austria
A FAST IMAGE-SCRAMBLE METHOD USING PUBLIC-KEY ENCRYPTION ....................................... ,3435
Osumu Wotanabe. Akiko Nakazaki and Hitoshi Kiya, ToLyo Metropolitan University, Jupun
COMPLIANT ENCRYPTION OF IPEGZOOO CODESTREAMS 3439
Yongdong Wu and Robert H. Deng, Institute /or Infocomm Research, Singapore
ROBUST PERCEPTUAL IMAGE HASHING VIA MATRIX INVARIANTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3443
Suleyrnon S.Kozot, University of Illinois, Urbana-Champaign, USA: Ramarathnam Venkatesun and M. Kivunc Milicuk, Microsofl
Research, USA
PROGRESSIVE PROTECTION OF JPEG2000 CODE-STREAMS 3447
Yongdong Wu,Di Ma and Robert H. Deng, Institute/or Infocomm Research, Singapore
OPTIMAL THRESHOLDING FOR KEY GENERATION BASED ON BIOMETRICS. .................................... 3451
Wende Zliang, Carnegie Mellon Universiry, USA; Yao-Jen Chang, lndusrrial Technology Research Institute, Taiwan; and Tsulrun Chen,
Cnmegie Mellon University, USA
SECURITY ANALYSIS FOR KEY GENERATION SYSTEMS USING FACE IMAGES 3455
Wende Zhung. Chn Zhang and Tsuhun Chen. Camegie Mellon University, USA
A PUBLIC-KEY AUTHENTICATION WATERMARKING FOR BINARY IMAGES. .................................... .3459
Hae Yong Kim, Universidude de Sao Paulo. Escola Politecnica, Brazil; and Ricardo L. De Q u e i r o ~Universidade de Bmsilio, Brazil
HIGH-CAPACITY DATA HIDING IN HALITONE IMAGES USING MINIMAL ERROR BIT SEARCHING . . . . . . . . . . . . . . . . ,3463
Soo-Chung Pei and Jing-Ming Guo, National Taiwan University, Taiwan

SESSION WP-PB: Image Segmentation: Clustering and Statistical Methods

IMAGE SEGMENTATION AS REGULARIZED CLUSTERING: A FULLY GLOBAL CURVE EVOLUTION METHOD 3467
Carlos Vazquez,Amar Miriche and Ismail Ben Ayed. INRS-EML Canada
AUTOMATIC DETERMINATION OF INTRINSIC CLUSTER NUMBER SEQUENCE IN SPECTRAL CLUSTERING
USING RANDOM WALK ON GRAPH .............................. 3471
Xin Zheng and Xueyin Lin, Institute of HCI and Media Integration, Tsinghuu University, China
AUTOMATIC CLASSIFICATION OF TEETH IN BITEWING DENTAL IMAGES ...................................... .3475
Molrammad Mukoor and M o l ~ m e Abdel-Mottaleh.
d Department of ECE, University of Miunii, USA

xli
EVOLUTIONARY GIBBS SAMPLER FOR IMAGE SEGMENTATION, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3479
Xino Wang and Hun Wung, Nmyrng Tecbnological University, Singopore
SEGMENTATION OF REGIONS IN JPEG COMPRESSED MEDICAL IMAGES . . . . . . . . . . . . . . . . . . . . . . . .3483
Praiiiod Singlr. Sclrool of Co,nputer Science and Engineering, University of New Soutli Wules, Aur-Ira ia
BACKGROUND DIFFERENCING TECHNIQUE FOR IMAGE SEGMENTATION BASED ON THE STATUS
OF REFERENCE PIXELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3487
Toshiyuki Yoshirla, Dept. Informarion Science, Futui University, Jopnn
A WAVELET DOMAIN HIERARCHICAL HIDDEN MARKOV MODEL ...............
%lien Ye rind Clreng-Clung La, Dept. of Compurer Science. Kenr Srure Univ
VISUALAITENTION BASED ROI MAPS FROM GAZE TRACKING D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3495
Antlion? Ngqen. Knurl Clrandran and Sridlu Sri~llairfm,Queensland 1Jniverriry of Technolog?. Aitslmliir
VARIABLE METRIC FOR BINARY VECTOR QUANTIZATION ........ 3499
Ismo K2rrkk.hinen and Pusi Ffirnti, Depamrienr of Co,rzpurer Scien
A HEURISTIC K-MEANS CLUSTERING ALGORITHM BY KERNEL PCA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,3503
Mantoo X u and Pasi Pr'2mfi. Universiq of Joensuu, Finland

SESSION WP-P9: Image Scanning, Display, and Printing I1

PERCEPTUAL QUALITY METRICS: EVALUATION OF INDIVIIXIAL COMPONENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3507
Henedicre Foiitffine,HaLbrr Snrirlane and Adricn Tlion~m,IKCC?N, F r w ~ e
COLOR VIDEO DEMOSAICKING VIA MOTION ESTIMATION . . . . . ................
Xiaolin Wii arid Lei Zlmng, McMnsrer Universir?, Conorlrr
LIGHT FIELD COMPRESSION BASED ON PREDICTION PROPAGATING AND WAVELET PACKET . . . . . . . . . . . . . . . . . . . ,3515
Xu Dong, Drri Qionglroi and X u Wenli, Deparrnrent of Automarion, Tsinghun Unioersity, Cltinn
EXPRESSION MORPHING FROM DISTANT VIEWPOINTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......... 3519
Tho Fu and Hassan Foroosh, Universiry of Central Floridrr, USA
MERGING RATIO IMAGES BASED REALISTIC OBJECT CLASS RE-RENDERING . . . . . . . . . . . . . . . . . . . . . 3523
Yun PIA,lnstirrire of Artijicial Intelligence and Rohorics(Al&R). Xi'an Jimtong Universiry Cliinn
CAMERA BASED EVALUATION OF PHOTOMETRIC COMPENSATION METHODS ON MULTIPROJECTOR DISPLAYS
Adiri Majrsndec University @California, Imine, USA
VIDEO QUALITY METRIC FOR LOW BITRATE COMPRESSED VIDEOS ........................ ......... ,3531
Ee Ping Ong. Xiaokang Yang. Weisi Lin. Zhongkang Lu and Susu Yao. Institutefor Infocornm Research, Singapore
A FINE-STRUCTURE I M A G W I D E O QUALITY MEASURE USING LOCAL STATISTICS. ..................... 3535
Kyungnaar Kim and Lorry Davis, Dept. of Computer Science, Univ. of Maryland, College Park USA
NO REFERENCE QUALITY ASSESSMENT FOR JPEC2OOO COMPRESSED IMAGES ............................... 3539
Hanghnng Tong, Tsinghua University, China; Mingjing Li,Hongjiartg Zhang, Micmsofr Research Asia, China; and Chasgsliui Zhang,
Tsingliua University. China
SEGMENTATION-DRIVENPERCEFIVAL QUALITY METRICS. ................................. ......... .3543
Andrea Cavallam, Queen Mary University of London, UK; and Stefan Winklec Genista Corporation, Swirierlan
HALFTONECONTONE CONVERSION USING NEURAL NETWORKS ................................. 3547
Win-BinHuong, Yen- Wei Lu. You-Hwang Kuo, Alvin WK Su and Wei-Chen Chang, Dept. of Cotnpurer Science and Information
Engineering. Narional Cheng Kung University, Taiwan

xlii