Research map / 3D vision / Gaussian splatting
WorldSplat: Gaussian-Centric Feed-Forward 4D Scene Generation for Autonomous Driving
Ziyue Zhu, Zhanqian Wu, Zhenxin Zhu, Lijun Zhou, Haiyang Sun, BING WANG, Kun Ma, Guang Chen, Hangjun Ye, Jin Xie, jian Yang
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 Gaussian renderings as condition to the enhanced diffusion model (vs reprojected boxes/sketches only) (ablation) | WorldSplat · nuScenes NVS +-2m | FVD: 260.07 -> 75.26 (A -> B, 3D Gs) | Improved |
| aggregate multi-frame 4D Gaussians instead of single-frame 3D Gaussians (ablation) | WorldSplat · nuScenes NVS +-2m | FVD: 75.26 -> 50.73 | Improved |
| add the enhanced diffusion refinement model (ablation) | WorldSplat · nuScenes NVS +-2m | FVD: 107.58 -> 47.41 | Improved |
| add mixed-quality augmentation when training the enhanced model (ablation) | WorldSplat · nuScenes NVS +-2m | FVD: 50.73 -> 47.41 | Improved |
| use WorldSplat instead of DiST-4D for novel views | WorldSplat · nuScenes NVS +-4m | FVD: 105.29 -> 64.07 | Improved |
| add first-frame guidance | WorldSplat · nuScenes video generation | FVD_multi: 74.13 -> 16.57 | Improved |
| add generated data to real data for StreamPETR training | StreamPETR · nuScenes detection | mAP: 34.5 -> 38.5 | Improved |
Closest papers by meaning
- DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation (CVPR 2025 · similarity 0.91)
- X-Scene: Large-Scale Driving Scene Generation with High Fidelity and Flexible Controllability (NeurIPS 2025 · similarity 0.90)
- Glad: A Streaming Scene Generator for Autonomous Driving (ICLR 2025 · similarity 0.89)
- Genesis: Multimodal Driving Scene Generation with Spatio-Temporal and Cross-Modal Consistency (NeurIPS 2025 · similarity 0.89)
- DriveScape: High-Resolution Driving Video Generation by Multi-View Feature Fusion (CVPR 2025 · similarity 0.89)
- GS-DiT: Advancing Video Generation with Dynamic 3D Gaussian Fields through Efficient Dense 3D Point Tracking (CVPR 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.
