Research map / Point clouds / Segmentation and detection
SuperLine3D: Self-Supervised Line Segmentation and Description for LiDAR Point Cloud
Xiangrui Zhao, Sheng Yang, Tianxin Huang, Jun Chen, Teng Ma, Mingyang Li, Yong Liu
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 |
|---|---|---|---|
| use skip-encoding stride 4 instead of 1 (ablation) | SuperLine3D · KITTI odometry 08-10 registration | recall: 96.51% -> 97.68% | Improved |
| use skip-encoding stride 4 instead of 2 (ablation) | SuperLine3D · KITTI odometry 08-10 registration | recall: 96.70% -> 97.68% | Improved |
| use skip-encoding stride 4 instead of 6 (ablation) | SuperLine3D · KITTI odometry 08-10 registration | recall: 64.23% -> 97.68% | Improved |
| increase descriptor dimension from 16 to 64 (ablation) | SuperLine3D · KITTI odometry 08-10 registration | recall: 87.56% -> 97.68% | Improved |
| increase descriptor dimension from 32 to 64 (ablation) | SuperLine3D · KITTI odometry 08-10 registration | recall: 95.28% -> 97.68% | Improved |
| increase descriptor dimension from 64 to 128 (ablation) | SuperLine3D · KITTI odometry 08-10 registration | recall: 97.68% -> 96.70% | No clear effect |
| replace point features (D3Feat) with line features | registration · KITTI odometry 08-10 registration | recall: 98.90% -> 97.68% | Hurt |
| replace point features (D3Feat) with line features | registration · KITTI odometry 08-10 registration | RTE mean (m): 0.088 -> 0.087 | No clear effect |
| replace point features (D3Feat) with line features on unseen dataset | registration · Apollo SouthBay SanJoseDowntown (trained on KITTI) | RTE mean (m): 0.079 -> 0.045 | Improved |
| replace point features (D3Feat) with line features on unseen dataset | registration · Apollo SouthBay SanJoseDowntown (trained on KITTI) | recall: 95.94% -> 93.84% | Hurt |
| replace point features (SpinNet) with line features on unseen dataset | registration · Apollo SouthBay SanJoseDowntown (trained on KITTI) | recall: 75.66% -> 93.84% | Improved |
| use scale-invariant features instead of vanilla DGCNN at fixed scale | segmentation model · synthetic data, original scale | accuracy / mIOU | 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.
- SpinNet — SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration (CVPR 2021) · better on 5 of 5 shared results
- HRegNet — HRegNet: A Hierarchical Network for Large-Scale Outdoor LiDAR Point Cloud Registration (ICCV 2021) · better on 3 of 3 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
- Large-Scale Point Cloud Semantic Segmentation With Superpoint Graphs (CVPR 2018 · similarity 0.89)
- Self-Supervised Learning of LiDAR 3D Point Clouds via 2D-3D Neural Calibration (TPAMI 2025 · similarity 0.89)
- LESS: Label-Efficient Semantic Segmentation for LiDAR Point Clouds (ECCV 2022 · similarity 0.89)
- Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data (CVPR 2022 · similarity 0.89)
- PSA-SSL: Pose and Size-aware Self-Supervised Learning on LiDAR Point Clouds (CVPR 2025 · similarity 0.87)
- Beyond the Label Itself: Latent Labels Enhance Semi-supervised Point Cloud Panoptic Segmentation (AAAI 2024 · similarity 0.87)
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.
