Research map / 3D vision / 3D generation
L3DR: 3D-aware LiDAR Diffusion and Rectification
Quan Liu, Xiaoqin Zhang, Ling Shao, Shijian Lu
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 MSE loss with Welsch loss in RRN training (ablation) | SPUNet RRN · SemanticKITTI | FSVD/FPVD: 26.3 -> 16.4 FSVD; 25.1 -> 12.1 FPVD | Improved |
| replace MSE loss with Welsch loss (ablation) | PTV3 RRN · SemanticKITTI | FSVD: 42.4 -> 17.7 | Improved |
| add segmentation map input to RRN (ablation) | SPUNet RRN · SemanticKITTI | FSVD: 16.4 -> 12.5 | Improved |
| use 3D UNet RRN instead of 2D image UNet (ablation) | L3DR · SemanticKITTI | FSVD/FPVD: 19.2 -> 16.4 FSVD; 16.4 -> 12.1 FPVD | Improved |
| swap SPUNet backbone for PTV3 (ablation) | RRN with Welsch · SemanticKITTI | FSVD: 16.4 -> 17.7 | No clear effect |
| add RRN rectification to LiDM | LiDM · KITTI360 unconditional | FSVD: 38.8 -> 35.8 | Improved |
| add RRN rectification to R2DM | R2DM · KITTI360 unconditional | JSD: 0.168 -> 0.165 | Improved |
| add RRN rectification to LiDM | LiDM · KITTI360 front-image-conditioned | FSVD: 32.5 -> 33.6 | Hurt |
| add RRN rectification to LiDM | LiDM · SemanticKITTI semantic-map-conditioned | FSVD: 20.2 -> 20.4 | No clear effect |
| add RRN rectification to LiDM | LiDM · Waymo semantic-conditioned | FSVD: 21.4 -> 18.3 | Improved |
| add RRN rectification to LiDM | LiDM · nuScenes semantic-conditioned | FSVD: 86.6 -> 81.3 | Improved |
Closest papers by meaning
- Towards Realistic Scene Generation with LiDAR Diffusion Models (CVPR 2024 · similarity 0.88)
- Scaling Diffusion Models to Real-World 3D LiDAR Scene Completion (CVPR 2024 · similarity 0.88)
- Blurred LiDAR for Sharper 3D: Robust Handheld 3D Scanning with Diffuse LiDAR and RGB (CVPR 2025 · similarity 0.87)
- Spiral: Semantic-Aware Progressive LiDAR Scene Generation and Understanding (NeurIPS 2025 · similarity 0.87)
- LSK3DNet: Towards Effective and Efficient 3D Perception with Large Sparse Kernels (CVPR 2024 · similarity 0.87)
- SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images (CVPR 2025 · similarity 0.85)
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.
