Actron3D：从视频中学习可操作的神经函数以实现可迁移的机器人操作
Actron3D: Learning Actionable Neural Functions from Videos for Transferable Robotic Manipulation
AI总结：
Actron3D通过神经可供性函数从少量单目视频中学习可迁移的6自由度操作技能，在仿真和真实世界中平均成功率提升14.9个百分点。
摘要：
我们提出了Actron3D，一个使机器人能够仅从少量单目、未标定、仅RGB的人类视频中获取可迁移的6自由度操作技能的框架。其核心是神经可供性函数（Neural Affordance Function），这是一种紧凑的以物体为中心的表示，将来自各种未标定视频的可操作线索——几何、视觉外观和可供性——提炼到一个轻量级神经网络中，形成操作技能的记忆库。在部署期间，我们采用一种流水线，通过由粗到细的优化检索相关的可供性函数并迁移精确的6自由度操作策略，这得益于对神经函数中编码的多模态特征的连续查询。在仿真和真实世界中的实验表明，Actron3D显著优于先前方法，在13个任务中平均成功率提高了14.9个百分点，而每个任务仅需2-3个演示视频。
英文摘要：
We present Actron3D, a framework that enables robots to acquire transferable 6-DoF manipulation skills from just a few monocular, uncalibrated, RGB-only human videos. At its core lies the Neural Affordance Function, a compact object-centric representation that distills actionable cues from diverse uncalibrated videos-geometry, visual appearance, and affordance-into a lightweight neural network, forming a memory bank of manipulation skills. During deployment, we adopt a pipeline that retrieves relevant affordance functions and transfers precise 6-DoF manipulation policies via coarse-to-fine optimization, enabled by continuous queries to the multimodal features encoded in the neural functions. Experiments in both simulation and the real world demonstrate that Actron3D significantly outperforms prior methods, achieving a 14.9 percentage point improvement in average success rate across 13 tasks while requiring only 2-3 demonstration videos per task.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
