NirAharon

Hello, the track.py is for the MOTChallenge benchmark. You can use one of the demo scripts for your own data. See the Readme.md file: https://github.com/NirAharon/BoT-SORT/blob/03a3a6851013abd54b95d863ec765c1437a63de8/README.md?plain=1#L211

Hello, Yes. Try to use our demo scripts: https://github.com/NirAharon/BoT-SORT/blob/a720e0490c6613dacccc0088f8c39c61725f1a63/README.md?plain=1#L213 Pass your YOLOX weights, exp files and video and it should work. If you trained your network for more than one...

May you provide more information? Maybe you need to adjust the tracker thresholds, did you change the --new_track_thresh argument according to your network scores? Did you check that the tracker...

It is because I didn't create any query folder for ranking evaluation. MOT17 and MOT20 aren't officially ReID datasets like Market1501. If you need this type of evaluation, you can...

Hi @EnzoLiang 1. The Kalman filter mean[0:1] (center x, center y) might be negative when the object is near the border of the image or leaving the image, is this...

Hi @EnzoLiang, sorry for the late response. Did you manage to adjust the parameters correctly?

Hi @Mr-Akbari, thank you for trying this. I will try to run BoT-SORT and BoT-SORT-ReID with the public detections, and get back to you with an answer.

Hi @deep-practice, please ask in English.

Hi @mikel-brostrom, this is not normal behavior. Please try to use the new --cmc-method I uploaded: sparseOptFlow. Please let me know if it was able to fix this problem.

Hi @C-monC and @mikel-brostrom, sorry for the late response I wasn't available for the last month. 1) BoT-SORT can achieve real-time. I created a c++ real-life implementation that I, unfortunately,...