Research map / Architectures and efficiency / RNNs and normalization
Towards Light-Weight and Real-Time Line Segment Detection
Read the paperCheck 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 |
|---|---|---|---|
| Add matching loss (ablation) | M-LSD-tiny trained with TP representation · Wireframe | sAP^10: 48.9 -> 52.2 (+3.3) | Helped |
| Add geometric loss (ablation) | M-LSD-tiny with matching loss · Wireframe | sAP^10: 52.2 -> 55.1 (+2.9) | Helped |
| Add SoL augmentation (ablation) | M-LSD-tiny with matching and geometric losses · Wireframe | sAP^10: 55.1 -> 58.0 (+2.9) | Helped |
| Add the M-LSD training schemes | L-CNN · Wireframe | sAP^10: 62.8 -> 63.7 | Helped |
| Add the M-LSD training schemes | HAWP · Wireframe | sAP^10: 66.5 -> 67.1 | Helped |
| Use M-LSD-tiny instead of TP-LSD-Lite at input 320 | Line segment detection on Wireframe · Wireframe, input 320 | FPS: 87.1 -> 200.8 | Helped |
Closest papers by meaning
- ScaleLSD: Scalable Deep Line Segment Detection Streamlined (CVPR 2025 · similarity 0.89)
- Improving Transformer Based Line Segment Detection with Matched Predicting and Re-ranking (AAAI 2025 · similarity 0.86)
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.82)
- GSENet:Global Semantic Enhancement Network for Lane Detection (AAAI 2024 · similarity 0.81)
- EdgeTAM: On-Device Track Anything Model (CVPR 2025 · similarity 0.81)
- MobileInst: Video Instance Segmentation on the Mobile (AAAI 2024 · similarity 0.80)
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.