Hi! I liked your approach and thanks for extending FairMOT for multi class tracking. However i would like to apply this approach on the 50 salads dataset. I am experimenting by annotating a few videos for both detection and tracking. Please let me know the steps i can take to run your code on my dataset. Also, please let me know the format for annotation files of detection and Ground truth for tracking.
@Ravsehajsinghpuri I'm sorry for the messy structure of the repo. I will refactor it if i have enough free time. I think you can refer to the issue #12 for details of the labeling.
Hi! I liked your approach and thanks for extending FairMOT for multi class tracking. However i would like to apply this approach on the 50 salads dataset. I am experimenting by annotating a few videos for both detection and tracking. Please let me know the steps i can take to run your code on my dataset. Also, please let me know the format for annotation files of detection and Ground truth for tracking.