Jinkun Cao

@AndrewGuo0930 Usually each issue is only for one question. For other problems, please initialize new issues with proper titles. Thanks.

@long2double @AndrewGuo0930 A simple multi-class tracking logic is supported in PR #548 , you may want to get some help there. We will support BDD100K in the near future. Thanks...

@AndrewGuo0930 You provided information lacks necessary knowledge for error tracing, such as the definition of your own datasets. Basically, I suggest to define your own customized dataset class by imitating...

To be precise, DET is designed for public-detection, which means different tracking algorithms share the detection results to be associated. If you are not targeting this goal, DET files are...

If no public detections are provided, you can define your own MOT datasets by imitating the support of DanceTrack dataset as in PR #543 .

@xulinxulin I recommend you to propose your expectation in the road of development (#73). We will schedule the development with taking the priority voted by the community into consideration.

Please check the evaluation server on CodaLab for [Youtube-VIS 2021](https://competitions.codalab.org/competitions/28988), which I believe is the currently only available entry for evaluation.

@gsygsygsy123 Have no idea if you have fixed this, but we could hardly do error tracing with the limited information your provide. I would recommend you to try with Pytorch...

I am considering to make support to some features in the short future. Let me know if some are already in process or your vote for their priority. 1. ~~Support...

Pose track is also a good to-do as a new task, baseline methods include PoseFlow[1]. The corresponding datasets are PoseTrack[2] and HiEve[3]. [1] "[Pose Flow: Efficient online pose tracking](https://arxiv.org/abs/1802.00977)" [2]...