Research map / Video / Video generation
Authentic 4D Driving Simulation with a Video Generation Model
Lening Wang, Wenzhao Zheng, Dalong Du, Yunpeng Zhang, Yilong Ren, Han Jiang, Zhiyong Cui, Haiyang Yu, Jie Zhou, Shanghang Zhang
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 point cloud fine alignment (ablation) | Stag · 3D reconstruction consistency | ATE: 0.957 -> 0.943 | Improved |
| add point cloud fine alignment (ablation) | Stag · 3D reconstruction consistency | Alignment error Ek: 5.901 -> 5.288 | Improved |
| add point cloud fine alignment (ablation) | Stag · view-change video consistency | Subject Consistency: 92.53% -> 93.71% | Improved |
| add the spatial-temporal block (ablation) | Stag · view-change video consistency | Subject Consistency: 93.71% -> 94.96% | Improved |
| add the spatial-temporal block (ablation) | Stag · time-change video consistency | Background Consistency: 94.91% -> 96.04% | Improved |
| add the spatial-temporal block (ablation) | Stag · 3D reconstruction consistency | ATE: 0.943 -> 0.943 | No clear effect |
| use Stag instead of S3Gaussian | Waymo-NOTR D32 · novel view synthesis | PSNR*: 22.92 -> 24.98 | Improved |
| use Stag instead of S3Gaussian | Waymo-NOTR D32 · novel view synthesis | SSIM*: 0.680 -> 0.644 | Hurt |
| use Stag instead of EmerNeRF | Waymo-NOTR S32 · novel view synthesis | PSNR: 28.89 -> 28.56 | Hurt |
| use Stag instead of StreetGaussian | Waymo Open Dataset multi-view · reconstruction | PSNR: 22.47 -> 26.03 | Improved |
| use Stag instead of ViewCrafter | frozen time generation · FVD | FVD: 218.6 -> 91.5 | Improved |
| use Stag instead of ViewCrafter | inference · total time | seconds: 119 -> 104.7 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MARS — MARS: Motion-Augmented RGB Stream for Action Recognition (CVPR 2019) · better on 2 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
- DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation (CVPR 2025 · similarity 0.92)
- DrivingSphere: Building a High-fidelity 4D World for Closed-loop Simulation (CVPR 2025 · similarity 0.90)
- X-Scene: Large-Scale Driving Scene Generation with High Fidelity and Flexible Controllability (NeurIPS 2025 · similarity 0.88)
- DriveScape: High-Resolution Driving Video Generation by Multi-View Feature Fusion (CVPR 2025 · similarity 0.88)
- Glad: A Streaming Scene Generator for Autonomous Driving (ICLR 2025 · similarity 0.87)
- UniScene: Unified Occupancy-centric Driving Scene Generation (CVPR 2025 · similarity 0.87)
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.
