Research map / Detection and segmentation / 3D detection and driving
V2V4Real: A Real-World Large-Scale Dataset for Vehicle-to-Vehicle Cooperative Perception
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 vehicle-to-vehicle cooperative perception instead of using only the ego vehicle | V2V4Real detection · V2V4Real cooperative 3D detection, Sync, Overall | AP@IoU=0.5: 39.8 -> 66.5 | Helped |
| Use intermediate feature fusion with CoBEVT rather than early fusion | V2V4Real detection · V2V4Real cooperative 3D detection, Sync, Overall | AP@IoU=0.5: 59.7 -> 66.5 | Helped |
| Use cooperative perception with communication delay rather than synchronous sharing | CoBEVT · V2V4Real cooperative 3D detection, Overall | AP@IoU=0.5/0.7: 66.5/36.0 -> 58.6/29.7 | Hurt |
| Use cooperative detections rather than single-vehicle detections for tracking | AB3Dmot tracking · V2V4Real cooperative tracking | AMOTA: 16.08 -> 32.12 | Helped |
| Add domain adaptation when transferring from OPV2V to V2V4Real | F-Cooper · OPV2V to V2V4Real | AP@IoU=0.5: 23.6 -> 37.3 (+13.7) | Helped |
| Add domain adaptation when transferring from OPV2V to V2V4Real | V2X-ViT · OPV2V to V2V4Real | AP@IoU=0.5: 27.4 -> 39.5 (+12.1) | Helped |
| Add domain adaptation when transferring from OPV2V to V2V4Real | CoBEVT · OPV2V to V2V4Real | AP@IoU=0.5: 32.6 -> 40.2 (+7.6) | Helped |
| Add domain adaptation when transferring from OPV2V to V2V4Real | AttFuse · OPV2V to V2V4Real | AP@IoU=0.5: 22.5 -> 23.4 (+0.9) | Helped |
| Add domain adaptation when transferring from OPV2V to V2V4Real | V2VNet · OPV2V to V2V4Real | AP@IoU=0.5: 23.2 -> 26.3 (+3.1) | Helped |
| Remove point-cloud rotation, flipping, and scaling augmentation (ablation) | CoBEVT · V2V4Real cooperative detection, Sync, Overall | AP@IoU=0.5: 66.5 -> 51.1(-15.4) | Hurt |
| Remove point-cloud rotation, flipping, and scaling augmentation (ablation) | Early Fusion · V2V4Real cooperative detection, Sync, Overall | AP@IoU=0.5: 59.7 -> 48.2(-11.5) | Hurt |
| Remove point-cloud rotation, flipping, and scaling augmentation (ablation) | V2VNet · V2V4Real cooperative detection, Sync, Overall | AP@IoU=0.5: 64.5 -> 49.0(-15.5) | Hurt |
Closest papers by meaning
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.92)
- 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 0.89)
- AGC-Drive: A Large-Scale Dataset for Real-World Aerial-Ground Collaboration in Driving Scenarios (NeurIPS 2025 · similarity 0.89)
- Transfer Your Perspective: Controllable 3D Generation from Any Viewpoint in a Driving Scene (CVPR 2025 · similarity 0.89)
- TUMTraf V2X Cooperative Perception Dataset (CVPR 2024 · similarity 0.89)
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.