ImDy: Human Inverse Dynamics from Imitated Observations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Xinpeng, Liang, Junxuan, Lin, Zili, Hou, Haowen, Li, Yong-Lu, Lu, Cewu
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929712483270656
author Liu, Xinpeng
Liang, Junxuan
Lin, Zili
Hou, Haowen
Li, Yong-Lu
Lu, Cewu
author_facet Liu, Xinpeng
Liang, Junxuan
Lin, Zili
Hou, Haowen
Li, Yong-Lu
Lu, Cewu
contents Inverse dynamics (ID), which aims at reproducing the driven torques from human kinematic observations, has been a critical tool for gait analysis. However, it is hindered from wider application to general motion due to its limited scalability. Conventional optimization-based ID requires expensive laboratory setups, restricting its availability. To alleviate this problem, we propose to exploit the recently progressive human motion imitation algorithms to learn human inverse dynamics in a data-driven manner. The key insight is that the human ID knowledge is implicitly possessed by motion imitators, though not directly applicable. In light of this, we devise an efficient data collection pipeline with state-of-the-art motion imitation algorithms and physics simulators, resulting in a large-scale human inverse dynamics benchmark as Imitated Dynamics (ImDy). ImDy contains over 150 hours of motion with joint torque and full-body ground reaction force data. With ImDy, we train a data-driven human inverse dynamics solver ImDyS(olver) in a fully supervised manner, which conducts ID and ground reaction force estimation simultaneously. Experiments on ImDy and real-world data demonstrate the impressive competency of ImDyS in human inverse dynamics and ground reaction force estimation. Moreover, the potential of ImDy(-S) as a fundamental motion analysis tool is exhibited with downstream applications. The project page is https://foruck.github.io/ImDy/.
format Preprint
id arxiv_https___arxiv_org_abs_2410_17610
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ImDy: Human Inverse Dynamics from Imitated Observations
Liu, Xinpeng
Liang, Junxuan
Lin, Zili
Hou, Haowen
Li, Yong-Lu
Lu, Cewu
Artificial Intelligence
Computer Vision and Pattern Recognition
Graphics
Robotics
Inverse dynamics (ID), which aims at reproducing the driven torques from human kinematic observations, has been a critical tool for gait analysis. However, it is hindered from wider application to general motion due to its limited scalability. Conventional optimization-based ID requires expensive laboratory setups, restricting its availability. To alleviate this problem, we propose to exploit the recently progressive human motion imitation algorithms to learn human inverse dynamics in a data-driven manner. The key insight is that the human ID knowledge is implicitly possessed by motion imitators, though not directly applicable. In light of this, we devise an efficient data collection pipeline with state-of-the-art motion imitation algorithms and physics simulators, resulting in a large-scale human inverse dynamics benchmark as Imitated Dynamics (ImDy). ImDy contains over 150 hours of motion with joint torque and full-body ground reaction force data. With ImDy, we train a data-driven human inverse dynamics solver ImDyS(olver) in a fully supervised manner, which conducts ID and ground reaction force estimation simultaneously. Experiments on ImDy and real-world data demonstrate the impressive competency of ImDyS in human inverse dynamics and ground reaction force estimation. Moreover, the potential of ImDy(-S) as a fundamental motion analysis tool is exhibited with downstream applications. The project page is https://foruck.github.io/ImDy/.
title ImDy: Human Inverse Dynamics from Imitated Observations
topic Artificial Intelligence
Computer Vision and Pattern Recognition
Graphics
Robotics
url https://arxiv.org/abs/2410.17610