Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.17303 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913136614834176 |
|---|---|
| author | Xu, Chenyi Wu, Yihao Yan, Liqi Yang, Chao Zhang, Jianhui Guan, Fangli Li, Pan |
| author_facet | Xu, Chenyi Wu, Yihao Yan, Liqi Yang, Chao Zhang, Jianhui Guan, Fangli Li, Pan |
| contents | Recovering a dynamic 3D scene from a long monocular video is crucial for dense geometry, camera motion, and temporal correspondence to remain consistent in a shared coordinate system. Existing methods face two key challenges: (1) feed-forward reconstruction models provide accurate local predictions but are limited to short clips, and (2) long-range trackers preserve correspondences without producing dense sequence-level reconstruction. This paper presents LongDPM, a novel overlap-aware framework for scalable long-range monocular dynamic reconstruction. First, LongDPM processes long videos in overlapping chunks, keeping inference memory bounded by the chunk length. Second, it connects chunk-local coordinate systems through confidence-weighted registration with static-aware overlap abstraction. Third, it associates dynamic identities across chunk boundaries and fuses matched trajectories to recover coherent long-range 3D motion. Experimental results demonstrate that LongDPM achieves superior long-range reconstruction and tracking performance, reducing dense tracking EPE over V-DPM on PointOdyssey, Kubric-F, and Kubric-G, while obtaining the best TUM-dynamics ATE for camera pose estimation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_17303 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LongDPM: Overlap-Aware 4D Reconstruction from Long Monocular Videos Xu, Chenyi Wu, Yihao Yan, Liqi Yang, Chao Zhang, Jianhui Guan, Fangli Li, Pan Computer Vision and Pattern Recognition Recovering a dynamic 3D scene from a long monocular video is crucial for dense geometry, camera motion, and temporal correspondence to remain consistent in a shared coordinate system. Existing methods face two key challenges: (1) feed-forward reconstruction models provide accurate local predictions but are limited to short clips, and (2) long-range trackers preserve correspondences without producing dense sequence-level reconstruction. This paper presents LongDPM, a novel overlap-aware framework for scalable long-range monocular dynamic reconstruction. First, LongDPM processes long videos in overlapping chunks, keeping inference memory bounded by the chunk length. Second, it connects chunk-local coordinate systems through confidence-weighted registration with static-aware overlap abstraction. Third, it associates dynamic identities across chunk boundaries and fuses matched trajectories to recover coherent long-range 3D motion. Experimental results demonstrate that LongDPM achieves superior long-range reconstruction and tracking performance, reducing dense tracking EPE over V-DPM on PointOdyssey, Kubric-F, and Kubric-G, while obtaining the best TUM-dynamics ATE for camera pose estimation. |
| title | LongDPM: Overlap-Aware 4D Reconstruction from Long Monocular Videos |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.17303 |