ScrewSplat: An End-to-End Method for Articulated Object Recognition
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909748517928960 |
|---|---|
| author | Kim, Seungyeon Ha, Junsu Kim, Young Hun Lee, Yonghyeon Park, Frank C. |
| author_facet | Kim, Seungyeon Ha, Junsu Kim, Young Hun Lee, Yonghyeon Park, Frank C. |
| contents | Articulated object recognition -- the task of identifying both the geometry and kinematic joints of objects with movable parts -- is essential for enabling robots to interact with everyday objects such as doors and laptops. However, existing approaches often rely on strong assumptions, such as a known number of articulated parts; require additional inputs, such as depth images; or involve complex intermediate steps that can introduce potential errors -- limiting their practicality in real-world settings. In this paper, we introduce ScrewSplat, a simple end-to-end method that operates solely on RGB observations. Our approach begins by randomly initializing screw axes, which are then iteratively optimized to recover the object's underlying kinematic structure. By integrating with Gaussian Splatting, we simultaneously reconstruct the 3D geometry and segment the object into rigid, movable parts. We demonstrate that our method achieves state-of-the-art recognition accuracy across a diverse set of articulated objects, and further enables zero-shot, text-guided manipulation using the recovered kinematic model. See the project website at: https://screwsplat.github.io. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_02146 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ScrewSplat: An End-to-End Method for Articulated Object Recognition Kim, Seungyeon Ha, Junsu Kim, Young Hun Lee, Yonghyeon Park, Frank C. Robotics Computer Vision and Pattern Recognition Articulated object recognition -- the task of identifying both the geometry and kinematic joints of objects with movable parts -- is essential for enabling robots to interact with everyday objects such as doors and laptops. However, existing approaches often rely on strong assumptions, such as a known number of articulated parts; require additional inputs, such as depth images; or involve complex intermediate steps that can introduce potential errors -- limiting their practicality in real-world settings. In this paper, we introduce ScrewSplat, a simple end-to-end method that operates solely on RGB observations. Our approach begins by randomly initializing screw axes, which are then iteratively optimized to recover the object's underlying kinematic structure. By integrating with Gaussian Splatting, we simultaneously reconstruct the 3D geometry and segment the object into rigid, movable parts. We demonstrate that our method achieves state-of-the-art recognition accuracy across a diverse set of articulated objects, and further enables zero-shot, text-guided manipulation using the recovered kinematic model. See the project website at: https://screwsplat.github.io. |
| title | ScrewSplat: An End-to-End Method for Articulated Object Recognition |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.02146 |