Research map / 3D vision / Gaussian splatting
DynamicVGGT: Learning Dynamic Point Maps for 4D Scene Reconstruction in Autonomous Driving
Zhuolin He, Jing Li, Guanghao Li, Xiaolei Chen, Jiacheng Tang, Siyang Zhang, Zhounan Jin, Feipeng Cai, Bin Li, Jian Pu, Jia Cai, Xiangyang Xue
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add temporal attention and future point head to VGGT (ablation) | VGGT · KITTI (monocular) | accuracy error: 1.489 -> 0.927 | Improved |
| add temporal attention and future point head to VGGT (ablation) | VGGT · KITTI (monocular) | completeness error: 0.690 -> 0.600 | Improved |
| add temporal attention and future point head to VGGT (ablation) | VGGT · Waymo val (3 cameras) | accuracy error: 4.635 -> 4.330 | Improved |
| add temporal attention and future point head to VGGT (ablation) | VGGT · Waymo val (3 cameras) | completeness error: 2.667 -> 2.939 | Hurt |
| add temporal attention and future point head to VGGT (ablation) | VGGT · KITTI (monocular) | normal consistency: 0.918 -> 0.915 | Hurt |
| add Dynamic 3DGS Head (stage 2) (ablation) | VGGT + TA & FPH · KITTI (monocular) | accuracy error: 0.927 -> 0.901 | Improved |
| add Dynamic 3DGS Head (stage 2) (ablation) | VGGT + TA & FPH · KITTI (monocular) | normal consistency: 0.915 -> 0.939 | Improved |
| add Dynamic 3DGS Head (stage 2) (ablation) | VGGT + TA & FPH · Waymo val (3 cameras) | accuracy error: 4.330 -> 4.021 | Improved |
| add Dynamic 3DGS Head (stage 2) (ablation) | VGGT + TA & FPH · Waymo val (3 cameras) | completeness error: 2.939 -> 2.390 | Improved |
| add Dynamic 3DGS Head (stage 2) (ablation) | VGGT + TA & FPH · KITTI (monocular) | completeness error: 0.600 -> 0.584 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- VGGT — VGGT: Visual Geometry Grounded Transformer (CVPR 2025) · better on 8 of 9 shared results
- GS-LRM — GS-LRM: Large Reconstruction Model for 3D Gaussian Splatting (ECCV 2024) · better on 0 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Flux4D: Flow-based Unsupervised 4D Reconstruction (NeurIPS 2025 · similarity 0.89)
- DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes (CVPR 2024 · similarity 0.88)
- SplatFlow: Self-Supervised Dynamic Gaussian Splatting in Neural Motion Flow Field for Autonomous Driving (CVPR 2025 · similarity 0.88)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.88)
- DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation (CVPR 2025 · similarity 0.88)
- DrivingRecon: Large 4D Gaussian Reconstruction Model For Autonomous Driving (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
