Research map / 3D vision / Gaussian splatting
Beyond Visual Reconstruction Quality: Object Perception-aware 3D Gaussian Splatting for Autonomous Driving
Renzhi Wang, Yuxiang Fu, Wuqi Wang, Haigen Min, Wei Feng, Lei Ma, Qing Guo
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 perception-aligned loss (ablation) | S3Gaussian · Waymo, YOLOv8 | mAP: 0.550 -> 0.593 | Improved |
| add perception-aligned loss (ablation) | OmniRe · Waymo, YOLOv8 | mAP: 0.489 -> 0.507 | Improved |
| add perception-aligned loss (ablation) | S3Gaussian · Waymo, Faster R-CNN black box | mAP: 0.171 -> 0.229 | Improved |
| add perception-aligned loss (ablation) | EMD(OmniRe) · Waymo, RT-DETR black box | mAP: 0.542 -> 0.542 | No clear effect |
| add perception-aligned loss (ablation) | EMD(S3G) · Waymo | SSIM: 0.923 -> 0.951 | Mixed |
| add object zone quality loss (ablation) | S3Gaussian · Waymo, YOLOv8 | mAP: 0.550 -> 0.672 | Improved |
| add object zone quality loss (ablation) | OmniRe · Waymo, YOLOv8 | mAP: 0.489 -> 0.545 | Improved |
| add object zone quality loss (ablation) | S3Gaussian · Waymo | Obj SSIM: 0.877 -> 0.921 | Improved |
| add object zone quality loss (ablation) | OmniRe · Waymo | PSNR: 33.77 -> 33.25 | Hurt |
| add both losses instead of object zone loss alone (ablation) | OmniRe · Waymo, YOLOv8 | mAP: 0.545 -> 0.609 | Improved |
| add both losses instead of object zone loss alone (ablation) | EMD(S3G) · Waymo, YOLOv8 | mAP: 0.601 -> 0.600 | No clear effect |
| add perception-aligned loss (ablation) | OmniRe · Waymo training time | time per 100 epochs (s): 34.84 -> 36.07 | Hurt |
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.
- OmniRe — OmniRe: Omni Urban Scene Reconstruction (ICLR 2025) · better on 4 of 5 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
- DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes (CVPR 2024 · similarity 0.90)
- Unifying Appearance Codes and Bilateral Grids for Driving Scene Gaussian Splatting (NeurIPS 2025 · similarity 0.89)
- DrivingForward: Feed-forward 3D Gaussian Splatting for Driving Scene Reconstruction from Flexible Surround-view Input (AAAI 2025 · similarity 0.89)
- Perceptual-GS: Scene-adaptive Perceptual Densification for Gaussian Splatting (ICML 2025 · similarity 0.88)
- SplatAD: Real-Time Lidar and Camera Rendering with 3D Gaussian Splatting for Autonomous Driving (CVPR 2025 · similarity 0.88)
- VR-Drive: Viewpoint-Robust End-to-End Driving with Feed-Forward 3D Gaussian Splatting (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.
