V2VSSC
收藏资源简介:
V2VSSC数据集是基于大规模V2V感知数据集OPV2V构建的首个V2V SSC基准,旨在通过车辆间通信技术解决自动驾驶中的感知遮挡问题。该数据集包含四个独特的模型和六个语义类别,涵盖道路、车辆、地形、建筑、植被和杆状物。通过CARLA模拟器精确获取的占用和语义细节,确保了数据的高准确性。V2VSSC数据集的应用领域主要集中在自动驾驶的3D场景理解,旨在提高自动驾驶车辆的安全性和导航精度。
The V2VSSC dataset is the first V2V SSC benchmark built on the large-scale V2V perception dataset OPV2V, aiming to solve the perception occlusion problem in autonomous driving through inter-vehicle communication technology. This dataset contains four unique models and six semantic categories, covering road, vehicle, terrain, building, vegetation, and pole. The accurate occupancy and semantic details obtained via the CARLA simulator ensure the high accuracy of the dataset. The application fields of the V2VSSC dataset mainly focus on 3D scene understanding for autonomous driving, with the purpose of improving the safety and navigation accuracy of autonomous vehicles.




