DrivingScene: A Multi-Task Online Feed-Forward 3D Gaussian Splatting Method for Dynamic Driving Scenes
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908618291412992 |
|---|---|
| author | Hou, Qirui Sun, Wenzhang Zeng, Chang Wang, Chunfeng Li, Hao Cui, Jianxun |
| author_facet | Hou, Qirui Sun, Wenzhang Zeng, Chang Wang, Chunfeng Li, Hao Cui, Jianxun |
| contents | Real-time, high-fidelity reconstruction of dynamic driving scenes is challenged by complex dynamics and sparse views, with prior methods struggling to balance quality and efficiency. We propose DrivingScene, an online, feed-forward framework that reconstructs 4D dynamic scenes from only two consecutive surround-view images. Our key innovation is a lightweight residual flow network that predicts the non-rigid motion of dynamic objects per camera on top of a learned static scene prior, explicitly modeling dynamics via scene flow. We also introduce a coarse-to-fine training paradigm that circumvents the instabilities common to end-to-end approaches. Experiments on nuScenes dataset show our image-only method simultaneously generates high-quality depth, scene flow, and 3D Gaussian point clouds online, significantly outperforming state-of-the-art methods in both dynamic reconstruction and novel view synthesis. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_24734 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DrivingScene: A Multi-Task Online Feed-Forward 3D Gaussian Splatting Method for Dynamic Driving Scenes Hou, Qirui Sun, Wenzhang Zeng, Chang Wang, Chunfeng Li, Hao Cui, Jianxun Computer Vision and Pattern Recognition Machine Learning Robotics Real-time, high-fidelity reconstruction of dynamic driving scenes is challenged by complex dynamics and sparse views, with prior methods struggling to balance quality and efficiency. We propose DrivingScene, an online, feed-forward framework that reconstructs 4D dynamic scenes from only two consecutive surround-view images. Our key innovation is a lightweight residual flow network that predicts the non-rigid motion of dynamic objects per camera on top of a learned static scene prior, explicitly modeling dynamics via scene flow. We also introduce a coarse-to-fine training paradigm that circumvents the instabilities common to end-to-end approaches. Experiments on nuScenes dataset show our image-only method simultaneously generates high-quality depth, scene flow, and 3D Gaussian point clouds online, significantly outperforming state-of-the-art methods in both dynamic reconstruction and novel view synthesis. |
| title | DrivingScene: A Multi-Task Online Feed-Forward 3D Gaussian Splatting Method for Dynamic Driving Scenes |
| topic | Computer Vision and Pattern Recognition Machine Learning Robotics |
| url | https://arxiv.org/abs/2510.24734 |