Research map / 3D vision / 3D generation
RoadPainter: Points Are Ideal Navigators for Topology transformER
Zhongxing Ma, Liang Shuang, Yongkun Wen, Weixin Lu, Guowei Wan
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 point-guided mask generation (PGM) to the baseline (ablation) | RoadPainter baseline · OpenLane-V2 subset_A | DET_l: 26.9 -> 28.1 | Improved |
| add point-mask fusion (PMF) on top of PGM (ablation) | RoadPainter · OpenLane-V2 subset_A | DET_l: 28.1 -> 30.7 | Improved |
| add SD map interaction module (ablation) | RoadPainter · OpenLane-V2 subset_A | TOP_ll: 7.9 -> 12.7 | Improved |
| add SD map interaction module (ablation) | RoadPainter · OpenLane-V2 subset_A | DET_t: 47.7 -> 47.1 | Hurt |
| add hybrid attention (masked cross-attention) to the decoder (ablation) | RoadPainter · OpenLane-V2 subset_A | DET_l: 29.6 -> 30.7 | Improved |
| add real-virtual separation self-attention (ablation) | RoadPainter · OpenLane-V2 subset_A | TOP_ll: 7.5 -> 7.9 | Improved |
| use segmentation both as auxiliary supervision and for centerline sampling instead of only one (ablation) | RoadPainter · OpenLane-V2 subset_A | DET_l: 29.4 -> 30.7 | Improved |
| use segmentation as auxiliary supervision versus detection only (ablation) | RoadPainter · OpenLane-V2 subset_A | DET_l: 26.9 -> 28.1 | Improved |
| RoadPainter vs TopoMLP (PETR architecture) on traffic element detection | RoadPainter · OpenLane-V2 subset_B | OLS: 38.7 -> 38.5 | Hurt |
Closest papers by meaning
- TopoCurve: Geometry-Aware Topology Reasoning via Bézier Curves in Autonomous Driving (NeurIPS 2026 · similarity 0.90)
- TopoLogic: An Interpretable Pipeline for Lane Topology Reasoning on Driving Scenes (NeurIPS 2024 · similarity 0.90)
- TopoPoint: Enhance Topology Reasoning via Endpoint Detection in Autonomous Driving (NeurIPS 2025 · similarity 0.89)
- TopoMLP: A Simple yet Strong Pipeline for Driving Topology Reasoning (ICLR 2024 · similarity 0.89)
- Bezier Everywhere All at Once: Learning Drivable Lanes as Bezier Graphs (CVPR 2024 · similarity 0.88)
- T2SG: Traffic Topology Scene Graph for Topology Reasoning in Autonomous Driving (CVPR 2025 · similarity 0.87)
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.
