Research map / Detection and segmentation / CNN features
CLRNet: Cross Layer Refinement Network for Lane Detection
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 CLRNet with DLA34 rather than CondLane with ResNet101 | Lane detection · CULane test | F1@50: 79.48 -> 80.47 | Helped |
| Use CLRNet with ResNet18 rather than LaneAF with DLA34 | Lane detection · LLAMAS valid | F1@50: 96.90 -> 96.96 | Helped |
| Use CLRNet with DLA34 rather than LaneAF with DLA34 | Lane detection · LLAMAS test | F1@50: 96.07 -> 96.12 | Helped |
| Use CLRNet with ResNet18 rather than CondLaneNet with ResNet101 | Lane detection · TuSimple test | F1: 97.24 -> 97.89 | Helped |
| Add Line IoU loss (ablation) | ResNet18 baseline · CULane test | mF1: 51.90 -> 52.80 | Helped |
| Add cross-layer refinement (ablation) | ResNet18 with Line IoU · CULane test | mF1: 52.80 -> 54.74 | Helped |
| Add ROIGather (ablation) | ResNet18 with Line IoU and refinement · CULane test | mF1: 54.74 -> 55.23 | Helped |
| Use progressive refinement from high-level R0 to lower-level R1 to R2 instead of the tested alternatives (ablation) | Cross-layer refinement · CULane test | mF1: 53.77 -> 54.74 | Helped |
| Replace smooth-L1 weight 0.5 with Line IoU loss weight 2 (ablation) | ResNet18 lane regressor · CULane test | mF1: 54.22 -> 55.23 | Helped |
| Increase smooth-L1 regression weight from 0.5 to 1.5 (ablation) | ResNet18 lane regressor · CULane test | mF1: 54.22 -> 52.68 | Hurt |
| Use only low-level single-stage refinement R2 instead of high-level R0 (ablation) | Cross-layer refinement · CULane test | F1@50: 78.27 -> 77.82 | Hurt |
Closest papers by meaning
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.89)
- GEONet: Global Enhancement and Optimization Network for Lane Detection (AAAI 2025 · similarity 0.88)
- A Hybrid Global-Local Perception Network for Lane Detection (AAAI 2024 · similarity 0.87)
- GSENet:Global Semantic Enhancement Network for Lane Detection (AAAI 2024 · similarity 0.87)
- A Siamese Transformer with Hierarchical Refinement for Lane Detection (NeurIPS 2024 · similarity 0.87)
- LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving (ICLR 2024 · similarity 0.85)
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.