HorizonRobotics/RoboTransfer

<h1 align="center"> RoboTransfer: Geometry-Consistent Video Diffusion for Robotic Visual Policy Transfer </h1>

<div align="center" class="authors"> Liu Liu, Xiaofeng Wang, Guosheng Zhao, Keyu Li, Wenkang Qin, Jiaxiong Qiu, Zheng Zhu, Guan Huang, Zhizhong Su </div>

<div align="center" style="line-height: 3;"> <a href="https://github.com/HorizonRobotics/RoboTransfer" target="blank" style="margin: 2px;"> <img alt="Code" src="https://img.shields.io/badge/Code-Github-blue" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://horizonrobotics.github.io/robotlab/robotransfer" target="blank" style="margin: 2px;"> <img alt="Project Page" src="https://img.shields.io/badge/🌐-ProjectPage-blue" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://arxiv.org/abs/2505.23171" target="blank" style="margin: 2px;"> <img alt="arXiv" src="https://img.shields.io/badge/📄-arXiv-b31b1b" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://youtu.be/dGXKtqDnm5Q" target="blank" style="margin: 2px;"> <img alt="Video" src="https://img.shields.io/badge/🎥-Video-red" style="display: inline-block; vertical-align: middle;"/> </a> <a href="https://mp.weixin.qq.com/s/c9-1HPBMHIy4oEwyKnsT7Q" target="_blank" style="margin: 2px;"> <img alt="中文介绍" src="https://img.shields.io/badge/中文介绍-07C160?logo=wechat&logoColor=white" style="display: inline-block; vertical-align: middle;"/> </a> </div>

<div align="center"> <img src="assets/pin.jpg" width="40%" alt="RoboTransfer"/></div>

🔍 Abstract

[image]

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

• —📐 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.
• —🔁 Cross-View Conditioning: Learns representations from multiple camera views with spatial correspondence.
• —🤖 Robotic Policy Transfer: Facilitates domain adaptation by generating synthetic training data in target domains.

📖 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 Li, Wenkang Qin, Jiaxiong Qiu, Zheng Zhu, Guan Huang, Zhizhong Su},
  journal={arXiv preprint arXiv:2505.23171},
  year={2025}
}