ReconDrive: Fast Feed-Forward 4D Gaussian Splatting for Autonomous Driving Scene Reconstruction
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914378622697472 |
|---|---|
| author | Yu, Haibao Xiao, Kuntao Wang, Jiahang Hao, Ruiyang Huang, Yuxin Hu, Guoran Qin, Haifang Jing, Bowen Bo, Yuntian Luo, Ping |
| author_facet | Yu, Haibao Xiao, Kuntao Wang, Jiahang Hao, Ruiyang Huang, Yuxin Hu, Guoran Qin, Haifang Jing, Bowen Bo, Yuntian Luo, Ping |
| contents | High-fidelity visual reconstruction and novel-view synthesis are essential for realistic closed-loop evaluation in autonomous driving. While 4D Gaussian Splatting (4DGS) offers a promising balance of accuracy and efficiency, existing per-scene optimization methods require costly iterative refinement, rendering them unscalable for extensive urban environments. Conversely, current feed-forward approaches often suffer from degraded photometric quality. To address these limitations, we propose ReconDrive, a feed-forward framework that leverages and extends the 3D foundation model VGGT for rapid, high-fidelity 4DGS generation. Our architecture introduces two core adaptations to tailor the foundation model to dynamic driving scenes: (1) Hybrid Gaussian Prediction Heads, which decouple the regression of spatial coordinates and appearance attributes to overcome the photometric deficiencies inherent in generalized foundation features; and (2) a Static-Dynamic 4D Composition strategy that explicitly captures temporal motion via velocity modeling to represent complex dynamic environments. Benchmarked on nuScenes, ReconDrive significantly outperforms existing feed-forward baselines in reconstruction, novel-view synthesis, and 3D perception. It achieves performance competitive with per-scene optimization while being orders of magnitude faster, providing a scalable and practical solution for realistic driving simulation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_07552 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ReconDrive: Fast Feed-Forward 4D Gaussian Splatting for Autonomous Driving Scene Reconstruction Yu, Haibao Xiao, Kuntao Wang, Jiahang Hao, Ruiyang Huang, Yuxin Hu, Guoran Qin, Haifang Jing, Bowen Bo, Yuntian Luo, Ping Computer Vision and Pattern Recognition Robotics High-fidelity visual reconstruction and novel-view synthesis are essential for realistic closed-loop evaluation in autonomous driving. While 4D Gaussian Splatting (4DGS) offers a promising balance of accuracy and efficiency, existing per-scene optimization methods require costly iterative refinement, rendering them unscalable for extensive urban environments. Conversely, current feed-forward approaches often suffer from degraded photometric quality. To address these limitations, we propose ReconDrive, a feed-forward framework that leverages and extends the 3D foundation model VGGT for rapid, high-fidelity 4DGS generation. Our architecture introduces two core adaptations to tailor the foundation model to dynamic driving scenes: (1) Hybrid Gaussian Prediction Heads, which decouple the regression of spatial coordinates and appearance attributes to overcome the photometric deficiencies inherent in generalized foundation features; and (2) a Static-Dynamic 4D Composition strategy that explicitly captures temporal motion via velocity modeling to represent complex dynamic environments. Benchmarked on nuScenes, ReconDrive significantly outperforms existing feed-forward baselines in reconstruction, novel-view synthesis, and 3D perception. It achieves performance competitive with per-scene optimization while being orders of magnitude faster, providing a scalable and practical solution for realistic driving simulation. |
| title | ReconDrive: Fast Feed-Forward 4D Gaussian Splatting for Autonomous Driving Scene Reconstruction |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2603.07552 |