SlideShare a Scribd company logo
DEEP LEARNING JP
[DL Papers]
“Human Dynamics from MonocularVideo
with Dynamic Camera Movements”
Presenter:Takahiro Maeda D1
(ToyotaTechnological Institute, Intelligent Information Media Lab)
http://deeplearning.jp/
目次
1. 書誌情報
2. 概要
3. 研究背景
4. 提案手法
5. 実験結果
6. 考察・所感
2
1. 書誌情報
紹介論文
題名:Human Dynamics from Monocular Video with Dynamic Camera Movements
出典:SIGGRAPH Asia 2021
著者:RI YU, HWANGPIL PARK, JEHEE LEE
所属:SNU
選書理由
今まで見過ごされていた3d human pose estimationの課題を
見事に解いていた.
3
結果
4
{人,カメラ}の動きが激しい単眼動画から人動作を移動量を含めて復元 Youtube link
2. 概要
• 激しく{人,カメラ}が移動する単眼動画から動作を復元する手法を提案した
• 研究背景: 激しいパン・チルト・ズームを含んだ動画は,
特徴点からカメラの自己位置推定が困難であり,
被写体の3次元空間の移動量を求めることが困難
→移動量が求められず姿勢推定が不完全
• アイデア: 物理法則(物理シミュレータ)からヒントを得て移動量を推定
• 手法: 2,3次元姿勢推定で得られた姿勢(関節角度)を
物理シミュレータ上で再現するように強化学習を行うことで,
3次元空間の移動量を推定 5
3. 研究背景
• 激しい動きを含んだ動画では,姿勢推定は移動量を正しく推定できない
– アニメーション技術やロボット動作獲得には適用不可
6
3次元姿勢推定結果
推定結果の重ね合わせ
移動量の推定失敗
• 人動作が物理法則に従うことを利用して姿勢などから移動量を推定する.
4. 提案手法: アイデア
7
姿勢推定結果
(root位置・回転な
し)
体の向き(2次元ベクトル),接触
(binary)
物理シミュレー
タ,
強化学習
による復元
復元された側転動作
4. 提案手法: ネットワーク
8
4. 提案手法: Contact Estimator, body orientation
9
2次元姿勢推定の結果から,体の向きを取
得
9フレーム分の2次元姿勢推定結果から
MLPにより接触を判定 Rempe et.al.[1]
学習データはCGシミュレータ上で作成
提案手法: Scene Arrangement
10
学習:大まかに与えたオブジェクトの位置,高さに外乱を加える
推論:後述する報酬を最大化するようにオブジェクトを配置(探
索)
4. 提案手法: 報酬設計
• 報酬
11
関節角 角速度 体の向き 接触 体中心への外力の制約
外力: RFC[2]において提案,
人体と3dモデルの差を修正す
る
推定した情報との一致
接触する
物体との距
離
接触する
物体との
方向の一致
接触タイミングから
予測される放物線との一
致
5. 実験結果
12
動きの激しい単眼動画から人動作を復元 Youtube link: https://www.youtube.com/watch?v=SOhgfAFN9eI
再掲
5. 実験結果: ablation
13
6. 考察・所感
• 従来の姿勢推定では無視されていた移動量を推定できる点は有用
– カメラ自己位置推定との比較実験が無いため疑問は残る
• 推定のために強化学習を行うことから,計算量が膨大
• 下準備が多いのは不便
– オブジェクトなどを用意した環境作成する
– カメラの動きが激しい想定なため,depth推定なので環境を再構成することも困難か
• 今後の展望
– 多人数が絡む動作(格闘技など)の動作再構成
– 水中や無重力下での動作再構成
– ロボットへのmotion transfer 14
Ad

More Related Content

Similar to [DL輪読会]Human Dynamics from Monocular Video with Dynamic Camera Movements (9)

Geotag Data Mining (メタサーベイ )
Geotag Data Mining (メタサーベイ )Geotag Data Mining (メタサーベイ )
Geotag Data Mining (メタサーベイ )
cvpaper. challenge
 
「解説資料」ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
「解説資料」ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation 「解説資料」ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
「解説資料」ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
Takumi Ohkuma
 
【DL輪読会】ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
【DL輪読会】ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation【DL輪読会】ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
【DL輪読会】ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
Deep Learning JP
 
[DL輪読会]"Omnimatte: Associating Objects and Their Effects in Video"
[DL輪読会]"Omnimatte: Associating Objects and Their Effects in Video"[DL輪読会]"Omnimatte: Associating Objects and Their Effects in Video"
[DL輪読会]"Omnimatte: Associating Objects and Their Effects in Video"
Deep Learning JP
 
20150930
2015093020150930
20150930
nlab_utokyo
 
【DL輪読会】Masked World Models for Visual Control
【DL輪読会】Masked World Models for Visual Control【DL輪読会】Masked World Models for Visual Control
【DL輪読会】Masked World Models for Visual Control
Deep Learning JP
 
Deeplearning lt.pdf
Deeplearning lt.pdfDeeplearning lt.pdf
Deeplearning lt.pdf
Deep Learning JP
 
ACCV2014参加報告
ACCV2014参加報告ACCV2014参加報告
ACCV2014参加報告
Yasutomo Kawanishi
 
[2019-03-14] JSPP19 深層学習による植物注釈タスクとPublic Cloud活用法
[2019-03-14] JSPP19 深層学習による植物注釈タスクとPublic Cloud活用法[2019-03-14] JSPP19 深層学習による植物注釈タスクとPublic Cloud活用法
[2019-03-14] JSPP19 深層学習による植物注釈タスクとPublic Cloud活用法
Eli Kaminuma
 
Geotag Data Mining (メタサーベイ )
Geotag Data Mining (メタサーベイ )Geotag Data Mining (メタサーベイ )
Geotag Data Mining (メタサーベイ )
cvpaper. challenge
 
「解説資料」ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
「解説資料」ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation 「解説資料」ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
「解説資料」ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
Takumi Ohkuma
 
【DL輪読会】ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
【DL輪読会】ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation【DL輪読会】ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
【DL輪読会】ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation
Deep Learning JP
 
[DL輪読会]"Omnimatte: Associating Objects and Their Effects in Video"
[DL輪読会]"Omnimatte: Associating Objects and Their Effects in Video"[DL輪読会]"Omnimatte: Associating Objects and Their Effects in Video"
[DL輪読会]"Omnimatte: Associating Objects and Their Effects in Video"
Deep Learning JP
 
【DL輪読会】Masked World Models for Visual Control
【DL輪読会】Masked World Models for Visual Control【DL輪読会】Masked World Models for Visual Control
【DL輪読会】Masked World Models for Visual Control
Deep Learning JP
 
[2019-03-14] JSPP19 深層学習による植物注釈タスクとPublic Cloud活用法
[2019-03-14] JSPP19 深層学習による植物注釈タスクとPublic Cloud活用法[2019-03-14] JSPP19 深層学習による植物注釈タスクとPublic Cloud活用法
[2019-03-14] JSPP19 深層学習による植物注釈タスクとPublic Cloud活用法
Eli Kaminuma
 

More from Deep Learning JP (20)

【DL輪読会】AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners
【DL輪読会】AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners【DL輪読会】AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners
【DL輪読会】AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners
Deep Learning JP
 
【DL輪読会】事前学習用データセットについて
【DL輪読会】事前学習用データセットについて【DL輪読会】事前学習用データセットについて
【DL輪読会】事前学習用データセットについて
Deep Learning JP
 
【DL輪読会】 "Learning to render novel views from wide-baseline stereo pairs." CVP...
【DL輪読会】 "Learning to render novel views from wide-baseline stereo pairs." CVP...【DL輪読会】 "Learning to render novel views from wide-baseline stereo pairs." CVP...
【DL輪読会】 "Learning to render novel views from wide-baseline stereo pairs." CVP...
Deep Learning JP
 
【DL輪読会】Zero-Shot Dual-Lens Super-Resolution
【DL輪読会】Zero-Shot Dual-Lens Super-Resolution【DL輪読会】Zero-Shot Dual-Lens Super-Resolution
【DL輪読会】Zero-Shot Dual-Lens Super-Resolution
Deep Learning JP
 
【DL輪読会】BloombergGPT: A Large Language Model for Finance arxiv
【DL輪読会】BloombergGPT: A Large Language Model for Finance arxiv【DL輪読会】BloombergGPT: A Large Language Model for Finance arxiv
【DL輪読会】BloombergGPT: A Large Language Model for Finance arxiv
Deep Learning JP
 
【DL輪読会】マルチモーダル LLM
【DL輪読会】マルチモーダル LLM【DL輪読会】マルチモーダル LLM
【DL輪読会】マルチモーダル LLM
Deep Learning JP
 
【 DL輪読会】ToolLLM: Facilitating Large Language Models to Master 16000+ Real-wo...
 【 DL輪読会】ToolLLM: Facilitating Large Language Models to Master 16000+ Real-wo... 【 DL輪読会】ToolLLM: Facilitating Large Language Models to Master 16000+ Real-wo...
【 DL輪読会】ToolLLM: Facilitating Large Language Models to Master 16000+ Real-wo...
Deep Learning JP
 
【DL輪読会】AnyLoc: Towards Universal Visual Place Recognition
【DL輪読会】AnyLoc: Towards Universal Visual Place Recognition【DL輪読会】AnyLoc: Towards Universal Visual Place Recognition
【DL輪読会】AnyLoc: Towards Universal Visual Place Recognition
Deep Learning JP
 
【DL輪読会】Can Neural Network Memorization Be Localized?
【DL輪読会】Can Neural Network Memorization Be Localized?【DL輪読会】Can Neural Network Memorization Be Localized?
【DL輪読会】Can Neural Network Memorization Be Localized?
Deep Learning JP
 
【DL輪読会】Hopfield network 関連研究について
【DL輪読会】Hopfield network 関連研究について【DL輪読会】Hopfield network 関連研究について
【DL輪読会】Hopfield network 関連研究について
Deep Learning JP
 
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
Deep Learning JP
 
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
Deep Learning JP
 
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
Deep Learning JP
 
【DL輪読会】"Language Instructed Reinforcement Learning for Human-AI Coordination "
【DL輪読会】"Language Instructed Reinforcement Learning  for Human-AI Coordination "【DL輪読会】"Language Instructed Reinforcement Learning  for Human-AI Coordination "
【DL輪読会】"Language Instructed Reinforcement Learning for Human-AI Coordination "
Deep Learning JP
 
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
Deep Learning JP
 
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
Deep Learning JP
 
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
Deep Learning JP
 
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
Deep Learning JP
 
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
Deep Learning JP
 
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
Deep Learning JP
 
【DL輪読会】AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners
【DL輪読会】AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners【DL輪読会】AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners
【DL輪読会】AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners
Deep Learning JP
 
【DL輪読会】事前学習用データセットについて
【DL輪読会】事前学習用データセットについて【DL輪読会】事前学習用データセットについて
【DL輪読会】事前学習用データセットについて
Deep Learning JP
 
【DL輪読会】 "Learning to render novel views from wide-baseline stereo pairs." CVP...
【DL輪読会】 "Learning to render novel views from wide-baseline stereo pairs." CVP...【DL輪読会】 "Learning to render novel views from wide-baseline stereo pairs." CVP...
【DL輪読会】 "Learning to render novel views from wide-baseline stereo pairs." CVP...
Deep Learning JP
 
【DL輪読会】Zero-Shot Dual-Lens Super-Resolution
【DL輪読会】Zero-Shot Dual-Lens Super-Resolution【DL輪読会】Zero-Shot Dual-Lens Super-Resolution
【DL輪読会】Zero-Shot Dual-Lens Super-Resolution
Deep Learning JP
 
【DL輪読会】BloombergGPT: A Large Language Model for Finance arxiv
【DL輪読会】BloombergGPT: A Large Language Model for Finance arxiv【DL輪読会】BloombergGPT: A Large Language Model for Finance arxiv
【DL輪読会】BloombergGPT: A Large Language Model for Finance arxiv
Deep Learning JP
 
【DL輪読会】マルチモーダル LLM
【DL輪読会】マルチモーダル LLM【DL輪読会】マルチモーダル LLM
【DL輪読会】マルチモーダル LLM
Deep Learning JP
 
【 DL輪読会】ToolLLM: Facilitating Large Language Models to Master 16000+ Real-wo...
 【 DL輪読会】ToolLLM: Facilitating Large Language Models to Master 16000+ Real-wo... 【 DL輪読会】ToolLLM: Facilitating Large Language Models to Master 16000+ Real-wo...
【 DL輪読会】ToolLLM: Facilitating Large Language Models to Master 16000+ Real-wo...
Deep Learning JP
 
【DL輪読会】AnyLoc: Towards Universal Visual Place Recognition
【DL輪読会】AnyLoc: Towards Universal Visual Place Recognition【DL輪読会】AnyLoc: Towards Universal Visual Place Recognition
【DL輪読会】AnyLoc: Towards Universal Visual Place Recognition
Deep Learning JP
 
【DL輪読会】Can Neural Network Memorization Be Localized?
【DL輪読会】Can Neural Network Memorization Be Localized?【DL輪読会】Can Neural Network Memorization Be Localized?
【DL輪読会】Can Neural Network Memorization Be Localized?
Deep Learning JP
 
【DL輪読会】Hopfield network 関連研究について
【DL輪読会】Hopfield network 関連研究について【DL輪読会】Hopfield network 関連研究について
【DL輪読会】Hopfield network 関連研究について
Deep Learning JP
 
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
Deep Learning JP
 
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
Deep Learning JP
 
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
Deep Learning JP
 
【DL輪読会】"Language Instructed Reinforcement Learning for Human-AI Coordination "
【DL輪読会】"Language Instructed Reinforcement Learning  for Human-AI Coordination "【DL輪読会】"Language Instructed Reinforcement Learning  for Human-AI Coordination "
【DL輪読会】"Language Instructed Reinforcement Learning for Human-AI Coordination "
Deep Learning JP
 
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
Deep Learning JP
 
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
Deep Learning JP
 
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
Deep Learning JP
 
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
Deep Learning JP
 
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
Deep Learning JP
 
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
Deep Learning JP
 
Ad

[DL輪読会]Human Dynamics from Monocular Video with Dynamic Camera Movements