Research map / 3D vision / 3D generation
S-NeRF++: Autonomous Driving Simulation via Neural Reconstruction and Generation
Yurui Chen, Junge Zhang, Ziyang Xie, Wenye Li, Feihu Zhang, Jiachen Lu, Li 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 pose and vehicle track refinement (ablation) | S-NeRF++ · nuScenes 2 scenes | PSNR: 28.21 -> 29.09 | Improved |
| add optimization-based insertion refinement (ablation) | object insertion · Waymo | FID: 73.8 -> 62.3 | Improved |
| use naive refinement instead of optimization (ablation) | object insertion · Waymo | FID: 67.1 -> 62.3 | Mixed |
| add confidence maps (ablation) | S-NeRF++ · nuScenes | PSNR: 23.15 -> 24.05 | Improved |
| add geometry confidence (ablation) | S-NeRF++ · nuScenes | PSNR: 23.88 -> 24.05 | Improved |
| vary geometry threshold tau (ablation) | S-NeRF++ · nuScenes | PSNR: 23.93 -> 23.97 | No clear effect |
| add view extrapolation with car insertion (ablation) | simulation data · nuScenes MV detection | AP: 26.5 -> 33.9 | Improved |
| add view interpolation to original views (ablation) | simulation data · nuScenes MV detection | AP pretrain: 22.9 -> 23.8 | No clear effect |
| add generated foreground assets (ablation) | foreground bank · Waymo segmentation | mIoU: 58.2 -> 62.8 | Improved |
| add reconstructed foreground assets (ablation) | foreground bank · nuScenes MV detection | AP: 28.7 -> 33.9 | Improved |
| joint training sim and real instead of pretrain then finetune (ablation) | DETR3D · nuScenes | Car AP: 45.7 -> 39.7 | Hurt |
| finetune sim-trained model with 500 real | DETR3D · nuScenes | Car AP: 33.9 -> 45.7 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- EmerNeRF — EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision (ICLR 2024) · better on 3 of 3 shared results
- S-NeRF — S-NeRF: Neural Radiance Fields for Street Views (ICLR 2023) · better on 3 of 3 shared results
- Zip-NeRF — Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields (ICCV 2023) · better on 3 of 3 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
- NeuRAD: Neural Rendering for Autonomous Driving (CVPR 2024 · similarity 0.90)
- NeRF-LiDAR: Generating Realistic LiDAR Point Clouds with Neural Radiance Fields (AAAI 2024 · similarity 0.88)
- Glad: A Streaming Scene Generator for Autonomous Driving (ICLR 2025 · similarity 0.87)
- DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation (CVPR 2025 · similarity 0.86)
- SplatAD: Real-Time Lidar and Camera Rendering with 3D Gaussian Splatting for Autonomous Driving (CVPR 2025 · similarity 0.86)
- PaReNeRF: Toward Fast Large-scale Dynamic NeRF with Patch-based Reference (CVPR 2024 · similarity 0.86)
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.
