论文/arXiv 2609.34674·1 个月前

HOI-Retarget：面向人-物交互的接触中心重定向

HOI-Retarget用接触中心重定向将人-物交互迁移到人形机器人，用于大规模运动数据生成。

AI 整理

• 本体 SDK 与控制
• 数据与采集
• 评测与安全

原文链接：HOI-Retarget: Contact-Centric Retargeting for Human-Object Interaction（arXiv 2609.34674 · 发布日期：2026-09-28）

HOI-Retarget: Contact-Centric Retargeting for Human-Object Interaction

HOI-Retarget用接触中心重定向将人-物交互迁移到人形机器人，用于大规模运动数据生成。

核心要点

• 人形机器人LfD扩展到人-物交互，受限于机器人兼容的交互参考不足。
• 方法以物体坐标系中的标注接触为目标，进行窗口化轨迹优化。
• 优化在机器人运动学限制下平衡身体跟踪、足部支撑与平滑性。
• 可跨物体尺寸扩增单次演示，吸收单目视频接触，并开源代码和数据集。

英文摘要 (Abstract)

Learning from demonstration (LfD) has enabled humanoid robots to acquire diverse whole-body skills, but extending this paradigm to human-object interaction (HOI) is limited by the availability of robot-compatible interaction references. We present HOI-Retarget, a contact-centric retargeting method that transfers HOI onto a humanoid robot for large-scale motion-data generation. Its windowed trajectory optimization uses every labeled contact as a target in the object frame, balancing body tracking, foot support and smoothness under the robot's kinematic limits. The method can augment a single demonstration across object sizes, absorb contacts reconstructed from monocular video, and extend to several robots manipulating one object. We publicly release the code and the retargeted motion dataset.

作者：Jihwan Shin, Adrià López Escoriza, Junzhe He, Heyrman, Matthias, Marco Hutter

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2609.34674：HOI-Retarget: Contact-Centric Retargeting for Human-Object Interaction访问 2026-10-21