Mix-Modality Person Re-Identification (MM-ReID) Dataset
收藏资源简介:
Mix-Modality Person Re-Identification (MM-ReID)数据集是由武汉科技大学和武汉大学联合创建的,旨在解决跨模态行人重识别问题。该数据集结合了可见光和红外图像,模拟了实际应用中白天和夜晚的行人图像识别需求。数据集的创建过程包括对现有数据集进行混合模态测试集的构建,并研究模态混合比例对性能的影响。该数据集主要应用于智能视频监控系统,旨在提高低光照条件下行人识别的准确性。
The Mix-Modality Person Re-Identification (MM-ReID) dataset was jointly developed by Wuhan University of Science and Technology and Wuhan University to address the cross-modality person re-identification task. This dataset integrates visible light and infrared images, simulating the practical requirements of pedestrian image recognition for both daytime and nighttime scenarios. The construction of the MM-ReID dataset involves building a mixed-modality test set based on existing datasets and investigating the effect of modality mixing ratios on model performance. This dataset is primarily utilized in intelligent video surveillance systems, with the aim of improving the accuracy of pedestrian recognition under low-light conditions.

- 1Mix-Modality Person Re-Identification: A New and Practical Paradigm武汉科技大学计算机科学与技术学院 · 2024年



