Even
当然，也可能是你某个地方两个分辨率不一致
@upsx You have misunderstood. We use multi-classifier rather than a classifier, as a result different classes use different IDs.
@upsx I think it may not have much difference from the view of MOT, but may cause confusion if all classes share the IDs.
@zhangmo123 训练数据很自由，可以组合公开数据集，也可以使用自己标注的数据集，只要组织成代码需要的形式即可，可以参考gen_labels_detrac_mcmot.py中的代码，方便的时候，我会放出一个小型自标注数据集的链接。
@s2244521 Just do ReID within each class respectively will be ok.
@s2244521 gen_mcmot_for_detect.py is used for generating labels for detections only(parse xml and convert to txt label), and the difference is track id all set to 0.
@s2244521 MOT16 is mainly for person tracking as far as i know, and it could hardly used for vehicle tracking training.
@ymRain 貌似MOT16数据集中的Vehicle数量很少，几乎不能当作MCMOT的训练集。
@ymRain 是的，第一列是class id， 具体的含义是: class_id, track_id, bbox_center_x, box_center_y, bbox_width, bbox_height, 祝顺利
@ymRain 方便分享一下你的数据集吗？方便的话我可以加到readme中去，让更多的人参与