Research map / Detection and segmentation / CNN features
Polynomial Regression Network for Variable-Number Lane Detection
Bingke Wang, Zilei Wang, Yixin Zhang
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 polynomial regression instead of semantic segmentation (ablation) | PRNet (BiSeNet-ResNet18) · TuSimple test | accuracy: 96.11% -> 97.18% | Improved |
| use polynomial regression instead of semantic segmentation (ablation) | PRNet (BiSeNet-ResNet18) · CULane test | F1: 70.7 -> 74.8 | Improved |
| use polynomial regression instead of semantic segmentation (ablation) | PRNet (ERFNet) · TuSimple test | accuracy: 95.92% -> 97.00% | Improved |
| use polynomial regression instead of semantic segmentation (ablation) | PRNet (ERFNet) · CULane test | F1: 73.6 -> 76.4 | Improved |
| raise polynomial order from 1 to 2 at piece height 16 (ablation) | PRNet · CULane val | F1: 76.85 -> 77.72 | Improved |
| raise polynomial order from 2 to 3 at piece height 16 (ablation) | PRNet · CULane val | F1: 77.72 -> 77.70 | No clear effect |
| raise polynomial order from 1 to 3 at piece height 64 (ablation) | PRNet · CULane val | F1: 76.66 -> 77.23 | Improved |
| reduce piece height from 64 to 16 at order 2 (ablation) | PRNet · CULane val | F1: 76.80 -> 77.72 | Improved |
| reduce piece height from 16 to 8 at order 2 (ablation) | PRNet · CULane val | F1: 77.72 -> 77.48 | Hurt |
| use BiSeNet backbone instead of ERFNet | PRNet · CULane test | F1: 76.4 -> 74.8 | Hurt |
| use BiSeNet backbone instead of ERFNet | PRNet · Run-time | FPS: 81 -> 110 | Improved |
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.
- FastDraw — FastDraw: Addressing the Long Tail of Lane Detection by Adapting a Sequential Prediction Network (CVPR 2019) · better on 1 of 1 shared result
- 3D-LaneNet — 3D-LaneNet: End-to-End 3D Multiple Lane Detection (ICCV 2019) · better on 1 of 1 shared result
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
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.87)
- LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving (ICLR 2024 · similarity 0.87)
- GEONet: Global Enhancement and Optimization Network for Lane Detection (AAAI 2025 · similarity 0.87)
- GSENet:Global Semantic Enhancement Network for Lane Detection (AAAI 2024 · similarity 0.86)
- GLane3D: Detecting Lanes with Graph of 3D Keypoints (CVPR 2025 · similarity 0.86)
- A Siamese Transformer with Hierarchical Refinement for Lane Detection (NeurIPS 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.
