OPV2V
Introduced 2021-09-16
OPV2V is a large-scale open simulated dataset for Vehicle-to-Vehicle perception. It contains over 70 interesting scenes, 11,464 frames, and 232,913 annotated 3D vehicle bounding boxes, collected from 8 towns in CARLA and a digital town of Culver City, Los Angeles.
Benchmarks
16k/AP@0.7@Default16k/AP@0.7@CulverCity16k/AP502D Classification/AP@0.7@Default2D Classification/AP@0.7@CulverCity2D Classification/AP502D Object Detection/AP@0.7@Default2D Object Detection/AP@0.7@CulverCity2D Object Detection/AP503D/AP@0.7@Default3D/AP@0.7@CulverCity3D/AP503D Object Detection/AP@0.7@Default3D Object Detection/AP@0.7@CulverCity3D Object Detection/AP50Object Detection/AP@0.7@DefaultObject Detection/AP@0.7@CulverCityObject Detection/AP50