Research map / 3D vision / Gaussian splatting
Targeted Structure Completion for Sparse-View 3D Reconstruction in Autonomous Driving
Guoqing Wang, Pin Tang, Xiangxuan Ren, Liping Hou, Chao Ma
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 |
|---|---|---|---|
| replace Omni-Scene voxel branch with targeted structure completion | Omni-Scene · nuScenes | latency (s) / PSNR: latency 0.088 -> 0.058; PSNR 24.27 -> 24.65 | Improved |
| add random structure completion to pixel-aligned + depth init (ablation) | FocusGS · nuScenes | PSNR: 23.14 -> 23.40 | Improved |
| add geometric ambiguity localization instead of random completion (ablation) | FocusGS · nuScenes | PSNR: 23.40 -> 24.65 | Improved |
| add depth initialization (ablation) | pixel-aligned baseline · nuScenes | LPIPS: 0.290 -> 0.320 | Hurt |
| depth-gradient mask instead of learned mask (ablation) | FocusGS · nuScenes | PSNR: 24.03 -> 24.65 | Improved |
| depth-gradient mask instead of RGB edge mask (ablation) | FocusGS · nuScenes | PSNR: 23.76 -> 24.65 | Improved |
| increase query Gaussians from 10k to 60k (ablation) | FocusGS · nuScenes | PSNR: 23.79 -> 24.65 | Improved |
| increase query Gaussians from 60k to 120k (ablation) | FocusGS · nuScenes | PSNR: 24.65 -> 24.67 | No clear effect |
| use 4 blocks instead of 1 (ablation) | FocusGS · nuScenes | PSNR: 23.76 -> 24.65 | Improved |
| use 6 blocks instead of 4 (ablation) | FocusGS · nuScenes | PSNR: 24.65 -> 24.56 | Hurt |
| add FocusGS completion to MVSplat | MVSplat · RealEstate10K | PSNR: 26.39 -> 27.02 | Improved |
| standalone FocusGS vs MVSplat | FocusGS · RealEstate10K | PSNR: 26.39 -> 26.32 (PSNR); 0.128 -> 0.123 (LPIPS) | Mixed |
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.
- Omni-Scene — Omni-Scene: Omni-Gaussian Representation for Ego-Centric Sparse-View Scene Reconstruction (CVPR 2025) · better on 9 of 9 shared results
- MVSplat — MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images (ECCV 2024) · better on 9 of 9 shared results
- DrivingForward — DrivingForward: Feed-forward 3D Gaussian Splatting for Driving Scene Reconstruction from Flexible Surround-view Input (AAAI 2025) · better on 4 of 4 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
- Tri-Perspective View Decomposition for Geometry-Aware Depth Completion (CVPR 2024 · similarity 0.87)
- PaSCo: Urban 3D Panoptic Scene Completion with Uncertainty Awareness (CVPR 2024 · similarity 0.87)
- DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes (CVPR 2024 · similarity 0.87)
- Not All Voxels Are Equal: Hardness-Aware Semantic Scene Completion with Self-Distillation (CVPR 2024 · similarity 0.87)
- Three Cars Approaching within 100m! Enhancing Distant Geometry by Tri-Axis Voxel Scanning for Camera-based Semantic Scene Completion (CVPR 2025 · similarity 0.87)
- ODG: Occupancy Prediction Using Dual Gaussians (NeurIPS 2025 · 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.
