VRAI
收藏资源简介:
VRAI数据集是由西北工业大学计算机科学与工程学院创建的大规模无人机(UAV)车辆再识别(ReID)数据集,包含13,022个车辆实例的137,613张图像。该数据集通过两架DJI Phantom4无人机在不同位置和高度(15m至80m)拍摄,以增加视角和姿态的多样性。数据集不仅标注了车辆的身份,还包括颜色、类型及特定车辆部位等详细属性,旨在支持细粒度识别和解决UAV视觉应用中的车辆再识别问题。
The VRAI dataset is a large-scale unmanned aerial vehicle (UAV)-based vehicle re-identification (ReID) dataset developed by the School of Computer Science and Engineering, Northwestern Polytechnical University. It comprises 137,613 images corresponding to 13,022 unique vehicle instances. This dataset was collected using two DJI Phantom4 UAVs at diverse locations and altitudes ranging from 15m to 80m, to enrich the diversity of viewpoints and vehicle poses. In addition to annotating vehicle identities, the dataset also includes detailed attributes such as vehicle color, type, and specific vehicle parts. It aims to support fine-grained recognition and address vehicle re-identification issues in UAV visual applications.

- 1Vehicle Re-identification in Aerial Imagery: Dataset and Approach西北工业大学计算机科学与工程学院 · 2019年



