SceneTracker: Long-term Scene Flow Estimation Network
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915299143450624 |
|---|---|
| author | Wang, Bo Li, Jian Yu, Yang Liu, Li Sun, Zhenping Hu, Dewen |
| author_facet | Wang, Bo Li, Jian Yu, Yang Liu, Li Sun, Zhenping Hu, Dewen |
| contents | Considering that scene flow estimation has the capability of the spatial domain to focus but lacks the coherence of the temporal domain, this study proposes long-term scene flow estimation (LSFE), a comprehensive task that can simultaneously capture the fine-grained and long-term 3D motion in an online manner. We introduce SceneTracker, the first LSFE network that adopts an iterative approach to approximate the optimal 3D trajectory. The network dynamically and simultaneously indexes and constructs appearance correlation and depth residual features. Transformers are then employed to explore and utilize long-range connections within and between trajectories. With detailed experiments, SceneTracker shows superior capabilities in addressing 3D spatial occlusion and depth noise interference, highly tailored to the needs of the LSFE task. We build a real-world evaluation dataset, LSFDriving, for the LSFE field and use it in experiments to further demonstrate the advantage of SceneTracker in generalization abilities. The code and data are available at https://github.com/wwsource/SceneTracker. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_19924 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SceneTracker: Long-term Scene Flow Estimation Network Wang, Bo Li, Jian Yu, Yang Liu, Li Sun, Zhenping Hu, Dewen Computer Vision and Pattern Recognition Considering that scene flow estimation has the capability of the spatial domain to focus but lacks the coherence of the temporal domain, this study proposes long-term scene flow estimation (LSFE), a comprehensive task that can simultaneously capture the fine-grained and long-term 3D motion in an online manner. We introduce SceneTracker, the first LSFE network that adopts an iterative approach to approximate the optimal 3D trajectory. The network dynamically and simultaneously indexes and constructs appearance correlation and depth residual features. Transformers are then employed to explore and utilize long-range connections within and between trajectories. With detailed experiments, SceneTracker shows superior capabilities in addressing 3D spatial occlusion and depth noise interference, highly tailored to the needs of the LSFE task. We build a real-world evaluation dataset, LSFDriving, for the LSFE field and use it in experiments to further demonstrate the advantage of SceneTracker in generalization abilities. The code and data are available at https://github.com/wwsource/SceneTracker. |
| title | SceneTracker: Long-term Scene Flow Estimation Network |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2403.19924 |