Research map / Detection and segmentation / 3D detection and driving
V2U4Real: A Real-world Large-scale Dataset for Vehicle-to-UAV Cooperative Perception
Weijia Li, Haoen Xiang, Tianxu Wang, Shuaibing Wu, Qiming Xia, Cheng Wang, Chenglu Wen
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 UAV and vehicle LiDAR with CoAlign instead of vehicle-only detection | PointPillars detector · V2U4Real cooperative detection, sync, IoU 0.5 | 3D AP: 27.53 -> 56.67 | Improved |
| use UAV-side single-agent detection instead of vehicle-side | PV-RCNN · V2U4Real, Vehicle class | 3D AP@IoU0.7: 36.23 -> 55.15 | Improved |
| use UAV-side single-agent detection instead of vehicle-side for cyclists | PointPillars · V2U4Real, Cyclist class | 3D AP@IoU0.25: 60.45 -> 57.33 | Hurt |
| introduce communication delay (async instead of sync) | Early Fusion · V2U4Real cooperative detection, IoU 0.5 | 3D AP: 51.31 -> 30.94 | Hurt |
| introduce communication delay (async instead of sync) | CoAlign · V2U4Real cooperative detection, IoU 0.5 | 3D AP: 56.67 -> 50.81 | Hurt |
| introduce communication delay (async instead of sync) | V2X-ViT · V2U4Real cooperative detection, IoU 0.5 | 3D AP: 45.51 -> 45.74 | No clear effect |
| add UAV cooperation at long range (50-100m) | DSRC · V2U4Real cooperative detection, sync, IoU 0.5 | 3D AP: 15.54 -> 33.26 | Improved |
| add cooperative fusion with CoAlign for tracking | AB3DMOT tracker · V2U4Real cooperative tracking | AMOTA: 11.73 -> 22.08 | Improved |
| use late fusion instead of intermediate fusion (CoAlign) | V2U cooperative detection · V2U4Real sync, IoU 0.5 | 3D AP: 56.67 -> 43.61 | Hurt |
Closest papers by meaning
- AGC-Drive: A Large-Scale Dataset for Real-World Aerial-Ground Collaboration in Driving Scenarios (NeurIPS 2025 · similarity 0.92)
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.89)
- UAV3D: A Large-scale 3D Perception Benchmark for Unmanned Aerial Vehicles (NeurIPS 2024 · similarity 0.89)
- DI-V2X: Learning Domain-Invariant Representation for Vehicle-Infrastructure Collaborative 3D Object Detection (AAAI 2024 · similarity 0.89)
- UrbanIng-V2X: A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception (NeurIPS 2025 · similarity 0.89)
- BEVSync: Asynchronous Data Alignment for Camera-based Vehicle-Infrastructure Cooperative Perception Under Uncertain Delays (AAAI 2025 · similarity 0.88)
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.
