Research map / Detection and segmentation / Open-set and few-shot detection
A Dataset for Lane Instance Segmentation in Urban Environments
Brook Roberts, Sebastian Kaltwang, Sina Samangooei, Mark Pender-Bare, Konstantinos Tertikas, John Redford
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 |
|---|---|---|---|
| train on the new lane dataset instead of KITTI | SegNet ROAD · ROAD cross-dataset average | IoU: 67.0 -> 84.0 | Improved |
| train on the new lane dataset instead of CityScapes | SegNet ROAD · ROAD cross-dataset average | IoU: 75.8 -> 84.0 | Improved |
| train on the new lane dataset instead of Mapillary | SegNet ROAD · ROAD cross-dataset average | IoU: 81.1 -> 84.0 | Improved |
| test on a different dataset than the training one (Ours to Mapillary) | SegNet ROAD · ROAD IoU | IoU: 95.0 -> 85.4 | Hurt |
| test on our data instead of own validation set (Mapillary-trained) | SegNet ROAD · ROAD IoU | IoU: 90.0 -> 85.4 | Hurt |
| test KITTI-trained model on CityScapes instead of CityScapes-trained | SegNet ROAD · ROAD IoU | IoU: 90.0 -> 60.4 | Hurt |
| evaluate EGO model on KITTI instead of own test set | SegNet EGO · EGO segmentation | IoU: 88.5 -> 61.2 | Hurt |
| train EGO model on KITTI (95 images) instead of our data, test on ours | SegNet EGO · EGO segmentation | IoU: 88.5 -> 39.2 | Hurt |
| annotate highway sequences instead of urban sequences | semi-automated annotation · annotation time per sequence | seconds: 361 -> 100 | Improved |
| annotate rural sequences instead of urban sequences | semi-automated annotation · annotation time per sequence | seconds: 361 -> 140 | Improved |
Closest papers by meaning
- ONCE-3DLanes: Building Monocular 3D Lane Detection (CVPR 2022 · similarity 0.90)
- LaneCorrect: Self-Supervised Lane Detection (IJCV 2025 · similarity 0.90)
- LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving (ICLR 2024 · similarity 0.88)
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.87)
- MWSIS: Multimodal Weakly Supervised Instance Segmentation with 2D Box Annotations for Autonomous Driving (AAAI 2024 · similarity 0.87)
- Weakly Supervised Semantic Segmentation for Driving Scenes (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.
