trinhdat15/UVH-26

UVH-26（城市视觉黑客松数据集）是一个大规模、针对印度交通的摄像头图像数据集，由印度科学院的AIM小组发布，用于智能交通系统和车辆检测研究。该数据集包含26,646张高分辨率（1080p）帧，采样自班加罗尔约2,800个安全城市CCTV摄像头，覆盖4周时间。图像通过一场全国范围的众包黑客松进行注释，涉及565名大学生，产生了约180万个边界框，涵盖14个细粒度车辆类别，这些类别代表了印度交通条件。为了捕捉不同注释共识水平，数据集包括两个独立的注释集：UVH-26-MV（基于每张图像多个注释者的多数投票计算的最终标签）和UVH-26-ST（使用STAPLE算法生成的标签，这是一种基于期望最大化的概率共识方法，以提高可靠性）。这两个版本共享相同的图像数据，但边界框共识逻辑不同。UVH-26 (Urban Vision Hackathon Dataset) is a large-scale, India-specific traffic-camera image dataset released by AIM @ IISc for research in intelligent transportation systems and vehicle detection. It contains 26,646 high-resolution (1080p) frames sampled from approximately 2,800 Bengaluru Safe City CCTV cameras over a 4-week period. Images were annotated through a nationwide crowdsourced hackathon involving 565 college students, producing approximately 1.8 million bounding boxes across 14 fine-grained vehicle classes representative of Indian traffic conditions. To capture different levels of annotation consensus, UVH-26 includes two separate annotation sets: UVH-26-MV (final labels computed via majority voting across multiple annotators per image) and UVH-26-ST (labels generated using the STAPLE algorithm, an Expectation–Maximization–based probabilistic consensus method for higher reliability). These versions share identical image data but differ in bounding box consensus logic.