Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Hu, Yucheng, Guo, Yanjiang, Wang, Pengchao, Chen, Xiaoyu, Wang, Yen-Jen, Zhang, Jianke, Sreenath, Koushil, Lu, Chaochao, Chen, Jianyu
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910926977892352
author Hu, Yucheng
Guo, Yanjiang
Wang, Pengchao
Chen, Xiaoyu
Wang, Yen-Jen
Zhang, Jianke
Sreenath, Koushil
Lu, Chaochao
Chen, Jianyu
author_facet Hu, Yucheng
Guo, Yanjiang
Wang, Pengchao
Chen, Xiaoyu
Wang, Yen-Jen
Zhang, Jianke
Sreenath, Koushil
Lu, Chaochao
Chen, Jianyu
contents Visual representations play a crucial role in developing generalist robotic policies. Previous vision encoders, typically pre-trained with single-image reconstruction or two-image contrastive learning, tend to capture static information, often neglecting the dynamic aspects vital for embodied tasks. Recently, video diffusion models (VDMs) demonstrate the ability to predict future frames and showcase a strong understanding of physical world. We hypothesize that VDMs inherently produce visual representations that encompass both current static information and predicted future dynamics, thereby providing valuable guidance for robot action learning. Based on this hypothesis, we propose the Video Prediction Policy (VPP), which learns implicit inverse dynamics model conditioned on predicted future representations inside VDMs. To predict more precise future, we fine-tune pre-trained video foundation model on robot datasets along with internet human manipulation data. In experiments, VPP achieves a 18.6\% relative improvement on the Calvin ABC-D generalization benchmark compared to the previous state-of-the-art, and demonstrates a 31.6\% increase in success rates for complex real-world dexterous manipulation tasks. Project page at https://video-prediction-policy.github.io
format Preprint
id arxiv_https___arxiv_org_abs_2412_14803
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations
Hu, Yucheng
Guo, Yanjiang
Wang, Pengchao
Chen, Xiaoyu
Wang, Yen-Jen
Zhang, Jianke
Sreenath, Koushil
Lu, Chaochao
Chen, Jianyu
Computer Vision and Pattern Recognition
Robotics
Visual representations play a crucial role in developing generalist robotic policies. Previous vision encoders, typically pre-trained with single-image reconstruction or two-image contrastive learning, tend to capture static information, often neglecting the dynamic aspects vital for embodied tasks. Recently, video diffusion models (VDMs) demonstrate the ability to predict future frames and showcase a strong understanding of physical world. We hypothesize that VDMs inherently produce visual representations that encompass both current static information and predicted future dynamics, thereby providing valuable guidance for robot action learning. Based on this hypothesis, we propose the Video Prediction Policy (VPP), which learns implicit inverse dynamics model conditioned on predicted future representations inside VDMs. To predict more precise future, we fine-tune pre-trained video foundation model on robot datasets along with internet human manipulation data. In experiments, VPP achieves a 18.6\% relative improvement on the Calvin ABC-D generalization benchmark compared to the previous state-of-the-art, and demonstrates a 31.6\% increase in success rates for complex real-world dexterous manipulation tasks. Project page at https://video-prediction-policy.github.io
title Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2412.14803