Research map / Detection and segmentation / 3D detection and driving
URScenes: A Multi-scenario Dataset for Unstructured Road Environments
Runsen Liu, Aizemaitijiang Baoerhan, Zhangyu Wang, Jie Wang, Jinghao Cui, Guizhen Yu, Songyue Yang, WanCheng Sun, Mingjun Tang, Zhanbo Hua, Wenwen Luo
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 |
|---|---|---|---|
| fuse camera with LiDAR in BEVFusion (ablation) | BEVFusion · URScenes cloudy subset | UDS (%): 75.8 -> 78.0 | Improved |
| fuse camera with LiDAR in FUTR3D (ablation) | FUTR3D · URScenes cloudy subset | UDS (%): 62.9 -> 75.7 | Improved |
| evaluate under dusty/foggy/snowy instead of clear weather | LiDAR-based detectors · URScenes detection | mAP (%): BEVFusion mAP 70.2 (cloudy) -> 16.7 (dusty) | Hurt |
| fuse LiDAR with camera in dusty scenes (ablation) | BEVFusion · URScenes dusty subset | UDS (%): 47.8 -> 44.3 | No clear effect |
| use LiDAR-camera fusion for 3D occupancy instead of unimodal | occupancy methods · URScenes occupancy | mIoU (%): FB-Occ camera 30.94 vs Co-Occ fusion 25.41 | Mixed |
| lower input resolution with efficient design (SparseOcc 704x256 vs SurroundOcc 1600x900) | occupancy models · URScenes occupancy | mIoU (%): 17.13 -> 26.83 | Improved |
| use end-to-end tracking vs two-stage tracking | MOT methods · URScenes tracking | AMOTA (%): ADA-Track 33.9 vs MCTrack 34.7 | No clear effect |
Closest papers by meaning
- RoScenes: A Large-scale Multi-view 3D Dataset for Roadside Perception (ECCV 2024 · similarity 0.89)
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (NeurIPS 2026 · similarity 0.88)
- UrbanIng-V2X: A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception (NeurIPS 2025 · similarity 0.87)
- RCooper: A Real-world Large-scale Dataset for Roadside Cooperative Perception (CVPR 2024 · similarity 0.87)
- TUMTraf V2X Cooperative Perception Dataset (CVPR 2024 · similarity 0.86)
- RoadSceneBench: A Lightweight Benchmark for Mid-Level Road Scene Understanding (CVPR 2026 · 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.
