CordViP: Correspondence-based Visuomotor Policy for Dexterous Manipulation in Real-World

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fu, Yankai, Feng, Qiuxuan, Chen, Ning, Zhou, Zichen, Liu, Mengzhen, Wu, Mingdong, Chen, Tianxing, Rong, Shanyu, Liu, Jiaming, Dong, Hao, Zhang, Shanghang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910920024784896
author Fu, Yankai
Feng, Qiuxuan
Chen, Ning
Zhou, Zichen
Liu, Mengzhen
Wu, Mingdong
Chen, Tianxing
Rong, Shanyu
Liu, Jiaming
Dong, Hao
Zhang, Shanghang
author_facet Fu, Yankai
Feng, Qiuxuan
Chen, Ning
Zhou, Zichen
Liu, Mengzhen
Wu, Mingdong
Chen, Tianxing
Rong, Shanyu
Liu, Jiaming
Dong, Hao
Zhang, Shanghang
contents Achieving human-level dexterity in robots is a key objective in the field of robotic manipulation. Recent advancements in 3D-based imitation learning have shown promising results, providing an effective pathway to achieve this goal. However, obtaining high-quality 3D representations presents two key problems: (1) the quality of point clouds captured by a single-view camera is significantly affected by factors such as camera resolution, positioning, and occlusions caused by the dexterous hand; (2) the global point clouds lack crucial contact information and spatial correspondences, which are necessary for fine-grained dexterous manipulation tasks. To eliminate these limitations, we propose CordViP, a novel framework that constructs and learns correspondences by leveraging the robust 6D pose estimation of objects and robot proprioception. Specifically, we first introduce the interaction-aware point clouds, which establish correspondences between the object and the hand. These point clouds are then used for our pre-training policy, where we also incorporate object-centric contact maps and hand-arm coordination information, effectively capturing both spatial and temporal dynamics. Our method demonstrates exceptional dexterous manipulation capabilities, achieving state-of-the-art performance in six real-world tasks, surpassing other baselines by a large margin. Experimental results also highlight the superior generalization and robustness of CordViP to different objects, viewpoints, and scenarios. Code and videos are available on https://aureleopku.github.io/CordViP.
format Preprint
id arxiv_https___arxiv_org_abs_2502_08449
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CordViP: Correspondence-based Visuomotor Policy for Dexterous Manipulation in Real-World
Fu, Yankai
Feng, Qiuxuan
Chen, Ning
Zhou, Zichen
Liu, Mengzhen
Wu, Mingdong
Chen, Tianxing
Rong, Shanyu
Liu, Jiaming
Dong, Hao
Zhang, Shanghang
Robotics
Artificial Intelligence
Achieving human-level dexterity in robots is a key objective in the field of robotic manipulation. Recent advancements in 3D-based imitation learning have shown promising results, providing an effective pathway to achieve this goal. However, obtaining high-quality 3D representations presents two key problems: (1) the quality of point clouds captured by a single-view camera is significantly affected by factors such as camera resolution, positioning, and occlusions caused by the dexterous hand; (2) the global point clouds lack crucial contact information and spatial correspondences, which are necessary for fine-grained dexterous manipulation tasks. To eliminate these limitations, we propose CordViP, a novel framework that constructs and learns correspondences by leveraging the robust 6D pose estimation of objects and robot proprioception. Specifically, we first introduce the interaction-aware point clouds, which establish correspondences between the object and the hand. These point clouds are then used for our pre-training policy, where we also incorporate object-centric contact maps and hand-arm coordination information, effectively capturing both spatial and temporal dynamics. Our method demonstrates exceptional dexterous manipulation capabilities, achieving state-of-the-art performance in six real-world tasks, surpassing other baselines by a large margin. Experimental results also highlight the superior generalization and robustness of CordViP to different objects, viewpoints, and scenarios. Code and videos are available on https://aureleopku.github.io/CordViP.
title CordViP: Correspondence-based Visuomotor Policy for Dexterous Manipulation in Real-World
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2502.08449