Research map / Graph learning / Knowledge and temporal graphs
VecRoad: Point-Based Iterative Graph Exploration for Road Graphs Extraction
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 RP-Net-Full instead of ImprovedConnectivity | Road graph extraction · RoadTracer dataset | J-F1: 55.21 -> 62.36 | Helped |
| Use RP-Net-Full instead of ImprovedConnectivity | Road graph extraction · RoadTracer dataset | APLS: 56.89 -> 61.14 | Helped |
| Use Res2Net as the backbone feature extractor (ablation) | RP-Net-Full · RoadTracer dataset | P-F1: 73.69 -> 72.56 | Hurt |
| Use Res2Net as the backbone feature extractor (ablation) | RP-Net-Full · RoadTracer dataset | J-F1: 62.36 -> 63.13 | Helped |
| Use Res2Net as the backbone feature extractor (ablation) | RP-Net-Full · RoadTracer dataset | APLS: 61.14 -> 64.59 | Helped |
| Add flexible step size (ablation) | RP-Net baseline · RoadTracer dataset | J-F1: 36.25 -> 43.83 | Helped |
| Add road segmentation cue to the flexible-step model (ablation) | RP-Net · RoadTracer dataset | P-F1: 56.42 -> 68.28 | Helped |
| Add junction segmentation cue to the flexible-step model (ablation) | RP-Net · RoadTracer dataset | J-F1: 43.83 -> 55.49 | Helped |
| Use road and junction segmentation cues together (ablation) | RP-Net with flexible step · RoadTracer dataset | APLS: 50.75 -> 57.28 | Helped |
| Add recurrent trajectory exploration (ablation) | RP-Net with flexible step and both cues · RoadTracer dataset | J-F1: 60.44 -> 62.36 | Helped |
| Use segmentation and post-processing instead of iterative graph exploration (ablation) | RP-Net · RoadTracer dataset | J-F1: 62.36 -> 50.16 | Hurt |
| Use segmentation and post-processing instead of iterative graph exploration (ablation) | RP-Net · RoadTracer dataset | APLS: 61.14 -> 49.68 | Hurt |
Closest papers by meaning
- Patched Line Segment Learning for Vector Road Mapping (AAAI 2024 · similarity 0.87)
- Towards Satellite Image Road Graph Extraction: A Global-Scale Dataset and A Novel Method (CVPR 2025 · similarity 0.85)
- Road Network Representation Learning with the Third Law of Geography (NeurIPS 2024 · similarity 0.84)
- Bezier Everywhere All at Once: Learning Drivable Lanes as Bezier Graphs (CVPR 2024 · similarity 0.84)
- LaneGraph2Seq: Lane Topology Extraction with Language Model via Vertex-Edge Encoding and Connectivity Enhancement (AAAI 2024 · similarity 0.83)
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.82)
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.