Research map / Graph learning / Graph self-supervised learning
RoadTagger: Robust Road Attribute Inference with Graph Neural Networks
Songtao He, Favyen Bastani, Satvat Jagwani, Edward Park, Sofiane Abbar, Mohammad Alizadeh, Hari Balakrishnan, Sanjay Chawla, Samuel Madden, Mohammad Amin Sadeghi
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 |
|---|---|---|---|
| Replace independent CNN image classification with end-to-end CNN and graph neural network inference | RoadTagger · Real-world dataset | lane accuracy: 71.8% -> 77.2% | Improved |
| Replace independent CNN image classification with end-to-end CNN and graph neural network inference | RoadTagger · Real-world dataset | road type accuracy: 89.1% -> 93.1% | Improved |
| Replace independent CNN image classification with end-to-end CNN and graph neural network inference | RoadTagger · Real-world dataset | absolute lane error (ALE): 22.2% | Improved |
| Add the road extraction graph instead of using only the original road network graph (ablation) | RoadTagger · Real-world dataset | lane accuracy: 74.0% -> 75.5% | Improved |
| Use the original, directionally decomposed road extraction, and auxiliary parallel-road graphs rather than only the original graph (ablation) | RoadTagger · Real-world dataset | lane accuracy: 74.0% -> 77.2% | Improved |
| Add random vertex dropout (ablation) | RoadTagger · Real-world dataset | lane accuracy: 74.7% -> 77.2% | Improved |
| Add random vertex dropout (ablation) | RoadTagger · Real-world dataset | road type accuracy: 92.7% -> 93.1% | Improved |
| Add graph Laplace regularization (ablation) | RoadTagger · Real-world dataset | road type accuracy: 90.8% -> 93.1% | Improved |
| Add graph Laplace regularization (ablation) | RoadTagger · Real-world dataset | lane accuracy: 76.5% -> 77.2% | Improved |
Closest papers by meaning
- Improved Road Connectivity by Joint Learning of Orientation and Segmentation (CVPR 2019 · similarity 0.87)
- TD-Road: Top-Down Road Network Extraction with Holistic Graph Construction (ECCV 2022 · similarity 0.87)
- Patched Line Segment Learning for Vector Road Mapping (AAAI 2024 · similarity 0.86)
- Road Network Representation Learning with the Third Law of Geography (NeurIPS 2024 · similarity 0.85)
- GEONet: Global Enhancement and Optimization Network for Lane Detection (AAAI 2025 · similarity 0.84)
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.84)
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.
