Research map / Point clouds / Point cloud classification
Flexible 3D Lane Detection by Hierarchical Shape Matching
Zhihao Guan, Ruixin Liu, Zejian Yuan, Ao Liu, Kun Tang, Tong Zhou, Erlong Li, Chao Zheng, Shuqi Mei
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 FHLD's hierarchical global and local predictions | 3D lane detection · RoadBEV, 10 cm | F1: 73.57 -> 77.57 | Improved |
| Use FHLD's hierarchical global and local predictions | 3D lane detection · RoadBEV, 30 cm | F1: 91.56 -> 93.74 | Improved |
| Use FHLD's hierarchical global and local predictions | 3D lane detection · sub-KCUD, 10 cm | F1: 72.27 -> 74.43 | Improved |
| Use FHLD's hierarchical global and local predictions | 3D lane detection · sub-KCUD, 30 cm | F1: 91.12 -> 92.20 | Improved |
| Add an adaptive reference axis for global curves rather than a fixed axis (ablation) | FHLD · RoadBEV | F1, 10 cm: 75.90(-1.67) -> 77.57 | Improved |
| Add local Gaussian shape matching rather than direct point regression (ablation) | FHLD · RoadBEV | F1, 10 cm: 75.62(-1.95) -> 77.57 | Improved |
| Add the local prediction branch to global curve predictions (ablation) | FHLD · RoadBEV | F1, 10 cm: 72.95(-4.62) -> 77.57 | Improved |
| Use an adaptive rather than fixed axis for global curves (ablation) | FHLD · hard-RoadBEV, parametric output, 10 cm | F1: 66.91 -> 71.36 | Improved |
| Use an adaptive rather than fixed axis for global curves (ablation) | FHLD · hard-RoadBEV, point set output, 10 cm | F1: 71.68 -> 74.67 | Improved |
| Vary anchor cell size from r=32 to r=16 (ablation) | FHLD · RoadBEV, 10 cm | F1: 77.57 -> 77.23 | Hurt |
| Vary anchor cell size from r=32 to r=24 (ablation) | FHLD · RoadBEV, 10 cm | F1: 77.57 -> 77.45 | Hurt |
| Vary anchor cell size from r=32 to r=40 (ablation) | FHLD · RoadBEV, 10 cm | F1: 77.57 -> 77.47 | Hurt |
Closest papers by meaning
- LaneCPP: Continuous 3D Lane Detection using Physical Priors (CVPR 2024 · similarity 0.89)
- Rethinking Lanes and Points in Complex Scenarios for Monocular 3D Lane Detection (CVPR 2025 · similarity 0.89)
- GLane3D: Detecting Lanes with Graph of 3D Keypoints (CVPR 2025 · similarity 0.88)
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.88)
- PVALane: Prior-Guided 3D Lane Detection with View-Agnostic Feature Alignment (AAAI 2024 · similarity 0.88)
- Bezier Everywhere All at Once: Learning Drivable Lanes as Bezier Graphs (CVPR 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.
