R+X: Retrieval and Execution from Everyday Human Videos
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909562501595136 |
|---|---|
| author | Papagiannis, Georgios Di Palo, Norman Vitiello, Pietro Johns, Edward |
| author_facet | Papagiannis, Georgios Di Palo, Norman Vitiello, Pietro Johns, Edward |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_12957 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | R+X: Retrieval and Execution from Everyday Human Videos Papagiannis, Georgios Di Palo, Norman Vitiello, Pietro Johns, Edward Robotics Machine Learning 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. |
| title | R+X: Retrieval and Execution from Everyday Human Videos |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2407.12957 |