arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.12957cs.ROcs.LG
收藏

R+X:从日常人类视频中检索与执行

R+X: Retrieval and Execution from Everyday Human Videos

Imperial College London(帝国理工学院)

Georgios Papagiannis, Norman Di Palo, Pietro Vitiello, Edward Johns

AI总结:

提出R+X框架,利用VLM从无标签第一人称人类视频中检索相关片段,并通过上下文模仿学习KAT使机器人无需训练即可立即执行语言指令技能,在家庭任务中表现优越。

摘要:

我们提出了R+X框架,使机器人能够从人类执行日常任务的长篇无标签第一人称视频中学习技能。给定人类的语言指令,R+X首先检索包含相关行为的短视频片段,然后通过将上下文模仿学习方法(KAT)条件化于该行为来执行技能。通过利用视觉语言模型(VLM)进行检索,R+X不需要对视频进行任何手动标注;通过利用上下文学习进行执行,机器人可以立即执行被指令的技能,而无需在检索到的视频上进行一段时间的训练。研究一系列日常家庭任务的实验表明,R+X成功地将无标签人类视频转化为鲁棒的机器人技能,并且R+X优于几种近期的替代方法。视频和代码可在https://www.robot-learning.uk/r-plus-x获取。

英文摘要:

We present R+X, a framework which enables robots to learn skills from long, unlabelled, first-person videos of humans performing everyday tasks. Given a language command from a human, R+X first retrieves short video clips containing relevant behaviour, and then executes the skill by conditioning an in-context imitation learning method (KAT) on this behaviour. By leveraging a Vision Language Model (VLM) for retrieval, R+X does not require any manual annotation of the videos, and by leveraging in-context learning for execution, robots can perform commanded skills immediately, without requiring a period of training on the retrieved videos. Experiments studying a range of everyday household tasks show that R+X succeeds at translating unlabelled human videos into robust robot skills, and that R+X outperforms several recent alternative methods. Videos and code are available at https://www.robot-learning.uk/r-plus-x.

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。

↑