SAM 3D: 3Dfy Anything in Images
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917096183562240 |
|---|---|
| author | SAM 3D Team Chen, Xingyu Chu, Fu-Jen Gleize, Pierre Liang, Kevin J Sax, Alexander Tang, Hao Wang, Weiyao Guo, Michelle Hardin, Thibaut Li, Xiang Lin, Aohan Liu, Jiawei Ma, Ziqi Sagar, Anushka Song, Bowen Wang, Xiaodong Yang, Jianing Zhang, Bowen Dollár, Piotr Gkioxari, Georgia Feiszli, Matt Malik, Jitendra |
| author_facet | SAM 3D Team Chen, Xingyu Chu, Fu-Jen Gleize, Pierre Liang, Kevin J Sax, Alexander Tang, Hao Wang, Weiyao Guo, Michelle Hardin, Thibaut Li, Xiang Lin, Aohan Liu, Jiawei Ma, Ziqi Sagar, Anushka Song, Bowen Wang, Xiaodong Yang, Jianing Zhang, Bowen Dollár, Piotr Gkioxari, Georgia Feiszli, Matt Malik, Jitendra |
| contents | We present SAM 3D, a generative model for visually grounded 3D object reconstruction, predicting geometry, texture, and layout from a single image. SAM 3D excels in natural images, where occlusion and scene clutter are common and visual recognition cues from context play a larger role. We achieve this with a human- and model-in-the-loop pipeline for annotating object shape, texture, and pose, providing visually grounded 3D reconstruction data at unprecedented scale. We learn from this data in a modern, multi-stage training framework that combines synthetic pretraining with real-world alignment, breaking the 3D "data barrier". We obtain significant gains over recent work, with at least a 5:1 win rate in human preference tests on real-world objects and scenes. We will release our code and model weights, an online demo, and a new challenging benchmark for in-the-wild 3D object reconstruction. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_16624 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SAM 3D: 3Dfy Anything in Images SAM 3D Team Chen, Xingyu Chu, Fu-Jen Gleize, Pierre Liang, Kevin J Sax, Alexander Tang, Hao Wang, Weiyao Guo, Michelle Hardin, Thibaut Li, Xiang Lin, Aohan Liu, Jiawei Ma, Ziqi Sagar, Anushka Song, Bowen Wang, Xiaodong Yang, Jianing Zhang, Bowen Dollár, Piotr Gkioxari, Georgia Feiszli, Matt Malik, Jitendra Computer Vision and Pattern Recognition Artificial Intelligence We present SAM 3D, a generative model for visually grounded 3D object reconstruction, predicting geometry, texture, and layout from a single image. SAM 3D excels in natural images, where occlusion and scene clutter are common and visual recognition cues from context play a larger role. We achieve this with a human- and model-in-the-loop pipeline for annotating object shape, texture, and pose, providing visually grounded 3D reconstruction data at unprecedented scale. We learn from this data in a modern, multi-stage training framework that combines synthetic pretraining with real-world alignment, breaking the 3D "data barrier". We obtain significant gains over recent work, with at least a 5:1 win rate in human preference tests on real-world objects and scenes. We will release our code and model weights, an online demo, and a new challenging benchmark for in-the-wild 3D object reconstruction. |
| title | SAM 3D: 3Dfy Anything in Images |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2511.16624 |