Research map / 3D vision / Gaussian splatting
AD-GS: Object-Aware B-Spline Gaussian Splatting for Self-Supervised Autonomous Driving
Jiawei Xu, Kai Deng, Zexin Fan, Shenlong Wang, 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 object and sky mask losses (ablation) | AD-GS · KITTI-75% | PSNR: 26.52 -> 26.98 | Improved |
| add flow and depth supervision (ablation) | AD-GS · KITTI-75% | PSNR: 26.98 -> 28.03 | Improved |
| add visibility and physically rigid regularization (ablation) | AD-GS · KITTI-75% | PSNR: 28.03 -> 29.16 | Improved |
| add trigonometric functions to B-spline motion (ablation) | AD-GS · Waymo | PSNR* (moving objects): 24.28 -> 26.65 | Improved |
| add B-spline curves to trigonometric functions (ablation) | AD-GS · Waymo | PSNR* (moving objects): 25.70 -> 26.65 | Improved |
| add bidirectional temporal visibility mask (ablation) | AD-GS · Waymo | PSNR* (moving objects): 26.65 -> 27.41 | Improved |
| add bidirectional temporal visibility mask (ablation) | AD-GS · Waymo | PSNR: 33.65 -> 33.91 | Improved |
| add flow supervision L_f (ablation) | AD-GS · Waymo | PSNR* (moving objects): 25.32 -> 27.41 | Improved |
| add flow supervision L_f (ablation) | AD-GS · KITTI-75% | PSNR: 28.84 -> 29.16 | Improved |
| use 3 frames per control point instead of 1 (ablation) | AD-GS · KITTI-75% | PSNR: 27.26 -> 29.16 | Improved |
| use 3 frames per control point instead of 4 (ablation) | AD-GS · KITTI-75% | PSNR: 29.11 -> 29.16 | No clear effect |
| use B-spline order 6 instead of 2 (ablation) | AD-GS · KITTI-75% | PSNR: 29.10 -> 29.16 | 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.
- Grid4D — Grid4D: 4D Decomposed Hash Encoding for High-Fidelity Dynamic Gaussian Splatting (NeurIPS 2024) · better on 5 of 5 shared results
- SUDS — SUDS: Scalable Urban Dynamic Scenes (CVPR 2023) · better on 3 of 3 shared results
- EmerNeRF — EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision (ICLR 2024) · 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
- DeSiRe-GS: 4D Street Gaussians for Static-Dynamic Decomposition and Surface Reconstruction for Urban Driving Scenes (CVPR 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)
- EGSRAL:An Enhanced 3D Gaussian Splatting Based Renderer with Automated Labeling for Large-Scale Driving Scene (AAAI 2025 · similarity 0.88)
- Unifying Appearance Codes and Bilateral Grids for Driving Scene Gaussian Splatting (NeurIPS 2025 · similarity 0.88)
- SplatAD: Real-Time Lidar and Camera Rendering with 3D Gaussian Splatting for Autonomous Driving (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.
