Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.14141 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910134549086208 |
|---|---|
| author | Chen, Lin-Zhuo Gao, Jian Chen, Yihang Cheng, Ka Leong Sun, Yipengjing Hu, Liangxiao Xue, Nan Zhu, Xing Shen, Yujun Yao, Yao Xu, Yinghao |
| author_facet | Chen, Lin-Zhuo Gao, Jian Chen, Yihang Cheng, Ka Leong Sun, Yipengjing Hu, Liangxiao Xue, Nan Zhu, Xing Shen, Yujun Yao, Yao Xu, Yinghao |
| contents | Streaming 3D reconstruction aims to recover 3D information, such as camera poses and point clouds, from a video stream, which necessitates geometric accuracy, temporal consistency, and computational efficiency. Motivated by the principles of Simultaneous Localization and Mapping (SLAM), we introduce LingBot-Map, a feed-forward 3D foundation model for reconstructing scenes from streaming data, built upon a geometric context transformer (GCT) architecture. A defining aspect of LingBot-Map lies in its carefully designed attention mechanism, which integrates an anchor context, a pose-reference window, and a trajectory memory to address coordinate grounding, dense geometric cues, and long-range drift correction, respectively. This design keeps the streaming state compact while retaining rich geometric context, enabling stable efficient inference at around 20 FPS on 518 x 378 resolution inputs over long sequences exceeding 10,000 frames. Extensive evaluations across a variety of benchmarks demonstrate that our approach achieves superior performance compared to both existing streaming and iterative optimization-based approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_14141 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Geometric Context Transformer for Streaming 3D Reconstruction Chen, Lin-Zhuo Gao, Jian Chen, Yihang Cheng, Ka Leong Sun, Yipengjing Hu, Liangxiao Xue, Nan Zhu, Xing Shen, Yujun Yao, Yao Xu, Yinghao Computer Vision and Pattern Recognition Streaming 3D reconstruction aims to recover 3D information, such as camera poses and point clouds, from a video stream, which necessitates geometric accuracy, temporal consistency, and computational efficiency. Motivated by the principles of Simultaneous Localization and Mapping (SLAM), we introduce LingBot-Map, a feed-forward 3D foundation model for reconstructing scenes from streaming data, built upon a geometric context transformer (GCT) architecture. A defining aspect of LingBot-Map lies in its carefully designed attention mechanism, which integrates an anchor context, a pose-reference window, and a trajectory memory to address coordinate grounding, dense geometric cues, and long-range drift correction, respectively. This design keeps the streaming state compact while retaining rich geometric context, enabling stable efficient inference at around 20 FPS on 518 x 378 resolution inputs over long sequences exceeding 10,000 frames. Extensive evaluations across a variety of benchmarks demonstrate that our approach achieves superior performance compared to both existing streaming and iterative optimization-based approaches. |
| title | Geometric Context Transformer for Streaming 3D Reconstruction |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.14141 |