Research map / Detection and segmentation / 3D detection and driving
Improving Online Lane Graph Extraction by Object-Lane Clustering
Yigit Baran Can, Alexander Liniger, Danda Pani Paudel, Luc Van Gool
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 object clustering loss (ablation) | lane graph model with DeepInteraction objects · NuScenes | M-F: 63.2 -> 64.2 | Improved |
| add object clustering loss (ablation) | lane graph model with BEVFusion objects · NuScenes | M-F: 63.4 -> 63.9 | Improved |
| add object clustering loss (ablation) | lane graph model with GT objects · NuScenes | M-F: 63.8 -> 64.2 | Improved |
| add object clustering loss (ablation) | lane graph model with CBGS objects · Argoverse | M-F: 67.6 -> 68.9 | Improved |
| add object clustering loss (ablation) | lane graph model with GT objects · Argoverse | Detect: 71.0 -> 71.0 | No clear effect |
| add object center refinement alone to the lane graph model (ablation) | lane graph model with DeepInteraction objects · NuScenes | M-F: 63.2 -> 63.2 | No clear effect |
| add refinement together with clustering loss (ablation) | lane graph model with DeepInteraction objects · NuScenes | M-F: 64.2 -> 64.8 | Improved |
| add lane graph supervision to object refinement training (ablation) | refinement module with DeepInteraction objects · NuScenes | mAP: 47.8 -> 58.6 | Improved |
| refine object centers vs original detector | DeepInteraction · NuScenes | mAP: 69.8 -> 58.9 | Hurt |
| test with different object detector than used in training (ablation) | lane graph model trained with BEVFusion objects · NuScenes | M-F: 64.9 -> 64.8 | No clear effect |
| test model trained on detections with ground-truth objects (ablation) | lane graph model trained with BEVFusion objects · NuScenes object clustering accuracy | accuracy (%): 90.3 -> 83.9 | Hurt |
Closest papers by meaning
- GLane3D: Detecting Lanes with Graph of 3D Keypoints (CVPR 2025 · similarity 0.88)
- LaneCPP: Continuous 3D Lane Detection using Physical Priors (CVPR 2024 · similarity 0.88)
- Bezier Everywhere All at Once: Learning Drivable Lanes as Bezier Graphs (CVPR 2024 · similarity 0.88)
- LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving (ICLR 2024 · similarity 0.87)
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.87)
- Rethinking Lanes and Points in Complex Scenarios for Monocular 3D Lane Detection (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.
