Syn4D: A Multiview Synthetic 4D Dataset
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910194796068864 |
|---|---|
| author | Jiang, Zeren Lan, Yushi Luo, Yihang Deng, Yufan Lai, Zihang Sucar, Edgar Rupprecht, Christian Laina, Iro Larlus, Diane Zheng, Chuanxia Vedaldi, Andrea |
| author_facet | Jiang, Zeren Lan, Yushi Luo, Yihang Deng, Yufan Lai, Zihang Sucar, Edgar Rupprecht, Christian Laina, Iro Larlus, Diane Zheng, Chuanxia Vedaldi, Andrea |
| contents | Dense 3D reconstruction and tracking of dynamic scenes from monocular video remains an important open challenge in computer vision. Progress in this area has been constrained by the scarcity of high-quality datasets with dense, complete, and accurate geometric annotations. To address this limitation, we introduce Syn4D, a multiview synthetic dataset of dynamic scenes that includes ground-truth camera motion, depth maps, dense tracking, and parametric human pose annotations. A key feature of Syn4D is the ability to unproject any pixel into 3D to any time and to any camera. We conduct extensive evaluations across multiple downstream tasks to demonstrate the utility and effectiveness of the proposed dataset, including 4D scene reconstruction, 3D point tracking, geometry-aware camera retargeting, and human pose estimation. The experimental results highlight Syn4D's potential to facilitate research in dynamic scene understanding and spatiotemporal modeling. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_05207 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Syn4D: A Multiview Synthetic 4D Dataset Jiang, Zeren Lan, Yushi Luo, Yihang Deng, Yufan Lai, Zihang Sucar, Edgar Rupprecht, Christian Laina, Iro Larlus, Diane Zheng, Chuanxia Vedaldi, Andrea Computer Vision and Pattern Recognition Dense 3D reconstruction and tracking of dynamic scenes from monocular video remains an important open challenge in computer vision. Progress in this area has been constrained by the scarcity of high-quality datasets with dense, complete, and accurate geometric annotations. To address this limitation, we introduce Syn4D, a multiview synthetic dataset of dynamic scenes that includes ground-truth camera motion, depth maps, dense tracking, and parametric human pose annotations. A key feature of Syn4D is the ability to unproject any pixel into 3D to any time and to any camera. We conduct extensive evaluations across multiple downstream tasks to demonstrate the utility and effectiveness of the proposed dataset, including 4D scene reconstruction, 3D point tracking, geometry-aware camera retargeting, and human pose estimation. The experimental results highlight Syn4D's potential to facilitate research in dynamic scene understanding and spatiotemporal modeling. |
| title | Syn4D: A Multiview Synthetic 4D Dataset |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.05207 |