Differentiable Robot Rendering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Ruoshi, Canberk, Alper, Song, Shuran, Vondrick, Carl
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910654678433792
author Liu, Ruoshi
Canberk, Alper
Song, Shuran
Vondrick, Carl
author_facet Liu, Ruoshi
Canberk, Alper
Song, Shuran
Vondrick, Carl
contents Vision foundation models trained on massive amounts of visual data have shown unprecedented reasoning and planning skills in open-world settings. A key challenge in applying them to robotic tasks is the modality gap between visual data and action data. We introduce differentiable robot rendering, a method allowing the visual appearance of a robot body to be directly differentiable with respect to its control parameters. Our model integrates a kinematics-aware deformable model and Gaussians Splatting and is compatible with any robot form factors and degrees of freedom. We demonstrate its capability and usage in applications including reconstruction of robot poses from images and controlling robots through vision language models. Quantitative and qualitative results show that our differentiable rendering model provides effective gradients for robotic control directly from pixels, setting the foundation for the future applications of vision foundation models in robotics.
format Preprint
id arxiv_https___arxiv_org_abs_2410_13851
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Differentiable Robot Rendering
Liu, Ruoshi
Canberk, Alper
Song, Shuran
Vondrick, Carl
Robotics
Computer Vision and Pattern Recognition
Graphics
Vision foundation models trained on massive amounts of visual data have shown unprecedented reasoning and planning skills in open-world settings. A key challenge in applying them to robotic tasks is the modality gap between visual data and action data. We introduce differentiable robot rendering, a method allowing the visual appearance of a robot body to be directly differentiable with respect to its control parameters. Our model integrates a kinematics-aware deformable model and Gaussians Splatting and is compatible with any robot form factors and degrees of freedom. We demonstrate its capability and usage in applications including reconstruction of robot poses from images and controlling robots through vision language models. Quantitative and qualitative results show that our differentiable rendering model provides effective gradients for robotic control directly from pixels, setting the foundation for the future applications of vision foundation models in robotics.
title Differentiable Robot Rendering
topic Robotics
Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2410.13851