Research map / Label-efficient and robust learning / Domain adaptation and generalization
Explicit Geometry-Reflectance Domain Shift Modeling for Robust LiDAR Segmentation in Adverse Weather
Longyu Yang, Shangbo Yuan, Lu Zhang, Jun Liu, Heng Tao Shen, Xiaofeng Zhu, Ping Hu
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 GRCNet geometry-reflectance collaboration to MinkNet | MinkNet · SemanticKITTI->SemanticSTF (DG) | mIoU: +18.1 mIoU | Improved |
| remove reflectance (geometry-only encoder) | MinkNet · SemanticSTF (DG) | mIoU: 24.4 -> 37.3 | Improved |
| add reflectance encoder fused by simple addition (ablation) | geometry-only encoder · SemanticSTF (DG) | mIoU: 37.3 -> 36.5 | Hurt |
| add complementarity-aware information constraint (ablation) | GB+RB addition · SemanticSTF (DG) | mIoU: 36.5 -> 38.6 | Improved |
| add local fusion (ablation) | GB+RB+CIC · SemanticSTF (DG) | mIoU: 38.6 -> 41.2 | Improved |
| add global fusion (ablation) | GB+RB+CIC+LF · SemanticSTF (DG) | mIoU: 41.2 -> 42.5 | Improved |
| use RMFC fusion instead of simple addition (ablation) | GRCNet · SemanticSTF (DG) | mIoU: 36.5 -> 42.5 | Improved |
| apply RMFC at shallower encoder stages too (stage 4 -> stages 4,3) (ablation) | GRCNet · SemanticSTF (DG) | mIoU: 42.5 -> 39.5 | Hurt |
| add TDDA data transfer to pseudo-label baseline (ablation) | GRCNet UDA · SemanticSTF (UDA) | mIoU: 42.3 -> 44.6 | Improved |
| use TDDA reflectance transfer alone (ablation) | pseudo-label baseline · SemanticSTF (UDA) | mIoU: 42.3 -> 42.5 | No clear effect |
| use range-view point drop transfer alone (ablation) | pseudo-label baseline · SemanticSTF (UDA) | mIoU: 42.3 -> 43.2 | Improved |
| replace all TDDA components with physics-based simulation (ablation) | GRCNet UDA · SemanticSTF (UDA) | mIoU: 44.6 -> 43.0 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- UniMix — UniMix: Towards Domain Adaptive and Generalizable LiDAR Semantic Segmentation in Adverse Weather (CVPR 2024) · better on 4 of 4 shared results
- CosMix — CoSMix: Compositional Semantic Mix for Domain Adaptation in 3D LiDAR Segmentation (ECCV 2022) · better on 1 of 1 shared result
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
- Towards Explicit Geometry-Reflectance Collaboration for Generalized LiDAR Segmentation in Adverse Weather (CVPR 2025 · similarity 0.96)
- UniMix: Towards Domain Adaptive and Generalizable LiDAR Semantic Segmentation in Adverse Weather (CVPR 2024 · similarity 0.89)
- No Thing, Nothing: Highlighting Safety-Critical Classes for Robust LiDAR Semantic Segmentation in Adverse Weather (CVPR 2025 · similarity 0.89)
- D^3CTTA: Domain-Dependent Decorrelation for Continual Test-Time Adaption of 3D LiDAR Segmentation (CVPR 2025 · similarity 0.87)
- Parsing All Adverse Scenes: Severity-Aware Semantic Segmentation with Mask-Enhanced Cross-Domain Consistency (AAAI 2024 · similarity 0.86)
- STGC-NeRF: Spatial-Temporal Geometric Consistency for LiDAR Neural Radiance Fields in Dynamic Scenes (AAAI 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.
