MSVR310
收藏资源简介:
MSVR310是一个高质量的多光谱车辆再识别基准数据集,由安徽省多模态认知计算重点实验室和安徽大学人工智能学院创建。该数据集包含310辆不同车辆,从广泛的视角、时间跨度和环境复杂性中收集,共计2087个样本。数据集通过RGB、近红外和热红外三种模态捕捉,旨在为多光谱车辆再识别提供一个全面的评估平台。数据集的创建过程考虑了多种环境干扰和车辆外观差异,以解决复杂光照环境下的车辆再识别问题。
MSVR310 is a high-quality multi-spectral vehicle re-identification benchmark dataset, created by the Anhui Provincial Key Laboratory of Multimodal Cognitive Computing and the School of Artificial Intelligence at Anhui University. This dataset contains 310 distinct vehicles, collected across a wide range of viewpoints, time spans and environmental complexities, with a total of 2087 samples. Captured using three modalities: RGB, near-infrared (NIR) and thermal infrared, it is designed to provide a comprehensive evaluation platform for multi-spectral vehicle re-identification. The development of this dataset takes into account various environmental disturbances and variations in vehicle appearances, aiming to solve the vehicle re-identification problem in complex lighting environments.




