Research map / Graph learning / Graph algorithms and clustering
DOGE: Differentiable Bézier Graph Optimization for Road Network Extraction
Jiahui Sun, Junran Lu, Jinhui Yin, Yishuo Xu, Yuanqi Li, Yanwen Guo
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 |
|---|---|---|---|
| add TopoAdapt topology refinement (ablation) | DOGE · SpaceNet | APLS: 69.45 -> 73.48 | Improved |
| add TopoAdapt topology refinement (ablation) | DOGE · SpaceNet | TOPO F1: 82.80 -> 84.58 | Improved |
| add overlap loss (ablation) | DOGE · SpaceNet | APLS: 65.95 -> 73.48 | Improved |
| add offset and spacing curve regularization (ablation) | DOGE · SpaceNet | APLS: 65.97 -> 73.48 | Improved |
| add all geometric priors (ablation) | DOGE · SpaceNet | TOPO F1: 74.58 -> 84.58 | Improved |
| add geometric priors and TopoAdapt to coverage-loss-only baseline (ablation) | DOGE · SpaceNet | APLS: 37.10 -> 73.48 | Improved |
| add geometric priors only (without TopoAdapt) to coverage-loss baseline (ablation) | L2-only baseline · SpaceNet | TOPO F1: 70.94 -> 82.80 | Improved |
| use Bezier graph with global optimization instead of SAMRoad++ polylines | SAMRoad++ · SpaceNet | TOPO F1: 81.57 -> 84.58 | Improved |
| use DOGE instead of SAMRoad++ | SAMRoad++ · City-Scale | APLS: 68.34 -> 70.24 | Improved |
| use DOGE instead of SAMRoad++ | SAMRoad++ · City-Scale | TOPO Precision: 88.39 -> 84.42 | Hurt |
| use curve-based Bezier graph instead of polylines | SAMRoad++ · City-Scale | Edges/km: 44.49 -> 14.00 | Improved |
Closest papers by meaning
- LineGraph2Road: Structural Graph Reasoning on Line Graphs for Road Network Extraction (ECCV 2026 · similarity 0.89)
- Bezier Everywhere All at Once: Learning Drivable Lanes as Bezier Graphs (CVPR 2024 · similarity 0.88)
- RoadTracer: Automatic Extraction of Road Networks From Aerial Images (CVPR 2018 · similarity 0.88)
- VecRoad: Point-Based Iterative Graph Exploration for Road Graphs Extraction (CVPR 2020 · similarity 0.88)
- Topology Preserving Local Road Network Estimation From Single Onboard Camera Image (CVPR 2022 · similarity 0.87)
- Patched Line Segment Learning for Vector Road Mapping (AAAI 2024 · similarity 0.86)
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.
