Research map / Point clouds / Segmentation and detection
Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the proposed cylindrical and asymmetrical 3D framework | LiDAR segmentation · SemanticKITTI test | mIoU: 66.4 -> 67.8 | Helped |
| Use the proposed cylindrical and asymmetrical 3D framework | LiDAR segmentation · nuScenes validation | mIoU: 72.2 -> 76.1 | Helped |
| Add cylindrical partition instead of cubic voxel partition (ablation) | 3D voxel baseline · SemanticKITTI validation | mIoU: 58.1 -> 61.0 | Helped |
| Add asymmetrical 3D convolution (ablation) | Cylinder model · SemanticKITTI validation | mIoU: 61.0 -> 63.8 | Helped |
| Add dimension-decomposition context modeling (ablation) | Cylinder + Asym-CNN · SemanticKITTI validation | mIoU: 63.8 -> 65.2 | Helped |
| Add point-wise refinement (ablation) | Cylinder + Asym-CNN + DDCM · SemanticKITTI validation | mIoU: 65.2 -> 65.9 | Helped |
| Replace a regular residual block with a 1D asymmetrical residual block (ablation) | 3D convolution network with cylindrical partition · SemanticKITTI validation | mIoU: 61.0 -> 62.0 | Helped |
| Include height in the asymmetrical residual block (ablation) | 1D-Asymmetrical residual block · SemanticKITTI validation | mIoU: 62.0 -> 63.8 | Helped |
| Add cylindrical partition and asymmetrical 3D convolution | SECOND · nuScenes 3D detection | mAP: 31.6 -> 36.4 | Helped |
| Add cylindrical partition and asymmetrical 3D convolution | SSN · nuScenes 3D detection | mAP: 46.3 -> 47.7 | Helped |
Closest papers by meaning
- Less Is More: Reducing Task and Model Complexity for 3D Point Cloud Semantic Segmentation (CVPR 2023 · similarity 0.92)
- Large-Scale Point Cloud Semantic Segmentation With Superpoint Graphs (CVPR 2018 · similarity 0.91)
- Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion (AAAI 2021 · similarity 0.90)
- Spherical Transformer for LiDAR-Based 3D Recognition (CVPR 2023 · similarity 0.90)
- PC-BEV: An Efficient Polar-Cartesian BEV Fusion Framework for LiDAR Semantic Segmentation (AAAI 2025 · similarity 0.88)
- 3D-AVS: LiDAR-based 3D Auto-Vocabulary Segmentation (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.