Research map / Point clouds / Segmentation and detection
RAPiD-Seg: Range-Aware Pointwise Distance Distribution Networks for 3D LiDAR Segmentation
Li Li, Hubert P. H. Shum, Toby P Breckon
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 RAPiD-Seg (RAPiD features, reflectivity, embedding, channel attention) to the baseline (ablation) | Minkowski-UNet34 (PCSeg) · SemanticKITTI val | mIoU: 70.04 -> 73.02 | Improved |
| add reflectivity fusion to RAPiD features (ablation) | RAPiD-Seg · SemanticKITTI val | mIoU: 71.21 -> 71.93 | Improved |
| use vanilla PDD features instead of RAPiD (ablation) | Minkowski-UNet34 · SemanticKITTI val | mIoU: 70.0 -> 66.2 | Hurt |
| use range-aware non-uniform k (10,7,5) instead of uniform k (ablation) | RAPiD · SemanticKITTI val | mIoU: 72.04 -> 73.02 | Improved |
| use RAPiD AE instead of ConvFNN for voxel embeddings (ablation) | RAPiD-Seg · SemanticKITTI val | mIoU: +0.24 mIoU | Improved |
| fuse R- and C-RAPiD-Seg features (ablation) | RAPiD-Seg · SemanticKITTI val | mIoU: 72.3 -> 73.0 | Improved |
| use channel-wise attention fusion instead of direct concatenation (ablation) | RAPiD-Seg · SemanticKITTI val | mIoU: +0.87 mIoU | Improved |
| swap backbone to Minkowski-UNet vs PTv2 vs Cylinder3D (ablation) | RAPiD-Seg · SemanticKITTI val | mIoU: 73.0 (Minkowski-UNet), 72.6 (PTv2), 69.8 (Cylinder3D) | Mixed |
| use RAPiD-Seg single-modal LiDAR versus multi-modal methods | RAPiD-Seg · SemanticKITTI test | mIoU: 75.2 -> 76.1 | Improved |
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.
- 2DPASS — 2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds (ECCV 2022) · 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
- SqueezeSegV3: Spatially-Adaptive Convolution for Efficient Point-Cloud Segmentation (ECCV 2020 · similarity 0.90)
- Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation (CVPR 2021 · similarity 0.90)
- Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-Based Perception (TPAMI 2021 · similarity 0.90)
- Rethinking Range View Representation for LiDAR Segmentation (ICCV 2023 · similarity 0.90)
- UniSeg: A Unified Multi-Modal LiDAR Segmentation Network and the OpenPCSeg Codebase (ICCV 2023 · similarity 0.90)
- PC-BEV: An Efficient Polar-Cartesian BEV Fusion Framework for LiDAR Semantic Segmentation (AAAI 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.
