DINOBot: Robot Manipulation via Retrieval and Alignment with Vision Foundation Models
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910337644625920 |
|---|---|
| author | Di Palo, Norman Johns, Edward |
| author_facet | Di Palo, Norman Johns, Edward |
| contents | We propose DINOBot, a novel imitation learning framework for robot manipulation, which leverages the image-level and pixel-level capabilities of features extracted from Vision Transformers trained with DINO. When interacting with a novel object, DINOBot first uses these features to retrieve the most visually similar object experienced during human demonstrations, and then uses this object to align its end-effector with the novel object to enable effective interaction. Through a series of real-world experiments on everyday tasks, we show that exploiting both the image-level and pixel-level properties of vision foundation models enables unprecedented learning efficiency and generalisation. Videos and code are available at https://www.robot-learning.uk/dinobot. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_13181 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DINOBot: Robot Manipulation via Retrieval and Alignment with Vision Foundation Models Di Palo, Norman Johns, Edward Robotics Machine Learning We propose DINOBot, a novel imitation learning framework for robot manipulation, which leverages the image-level and pixel-level capabilities of features extracted from Vision Transformers trained with DINO. When interacting with a novel object, DINOBot first uses these features to retrieve the most visually similar object experienced during human demonstrations, and then uses this object to align its end-effector with the novel object to enable effective interaction. Through a series of real-world experiments on everyday tasks, we show that exploiting both the image-level and pixel-level properties of vision foundation models enables unprecedented learning efficiency and generalisation. Videos and code are available at https://www.robot-learning.uk/dinobot. |
| title | DINOBot: Robot Manipulation via Retrieval and Alignment with Vision Foundation Models |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2402.13181 |