RoboTransfer / README.md
nemo04's picture
Update README.md
dd0c1a4 verified
|
raw
history blame
4.12 kB
metadata
license: apache-2.0
library_name: diffusers

RoboTransfer: Geometry-Consistent Video Diffusion for Robotic Visual Policy Transfer

Liu Liu, Xiaofeng Wang, Guosheng Zhao, Keyu Li, Wenkang Qin, Jiaxiong Qiu, Zheng Zhu, Guan Huang, Zhizhong Su
RoboTransfer Overview

The RoboTransfer framework integrates multi-view geometry and video diffusion, enabling controllable and geometry-consistent robotic video synthesis for policy transfer.


πŸ” Abstract

RoboTransfer is a novel diffusion-based video generation framework tailored for robotic visual policy transfer. Unlike conventional approaches, RoboTransfer introduces geometry-aware synthesis by injecting depth and normal priors, ensuring multi-view consistency across dynamic robotic scenes. The method further supports explicit control over scene components, such as background editing, object identity swapping, and motion specification, offering a fine-grained video generation pipeline that benefits embodied learning.


🧠 Key Features

  • πŸ“ Geometry-Consistent Diffusion: Injects global 3D cues (depth, normal) and cross-view interactions for multi-view realism.
  • 🧩 Scene Component Control: Enables manipulation of object attributes (pose, identity) and background features.
  • πŸ” Cross-View Conditioning: Learns representations from multiple camera views with spatial correspondence.
  • πŸ€– Robotic Policy Transfer: Facilitates domain adaptation by generating synthetic training data in target domains.

πŸ“¦ Resources


πŸ“Έ Framework Overview

RoboTransfer Pipeline

The overall architecture includes view-specific encoding, geometry injection, diffusion denoising with spatial constraints, and component-level editing modules. Our system enables compositional control over scene dynamics while preserving physical and geometric consistency.


πŸ“– BibTeX

@article{liu2025robotransfer,
  title={RoboTransfer: Geometry-Consistent Video Diffusion for Robotic Visual Policy Transfer},
  author={Liu, Liu and Wang, Xiaofeng and Zhao, Guosheng and Li, Keyu and Qin, Wenkang and Qiu, Jiaxiong and Zhu, Zheng and Huang, Guan and Su, Zhizhong},
  journal={arXiv preprint arXiv:2505.23171},
  year={2025}
}