Data Preprocessing Issue

[Realdr4g0n]

Thank you for sharing a great repository.

I suspect that there may be a problem with preprocessing the dataset. When creating det_db_0.json, which extracts the bounding boxes in advance through the BytetrackInference repository, I found that the json does not contain all image information due to the sampler.

If I am approaching this incorrectly, please let me know.

[zyayoung]

Could you provide more details about the sampler? My guess is that you are using a batch size > 1, which is not supported in ByteTrack.

[Realdr4g0n]

When running the ByteTrackerInference code without any modifications, the entire image assigned to the path was not output as a JSON file. However, when the sampler was removed from this code and inference was run, the bounding boxes for all images were output as JSON.

In this image, det_db_motrv2.json was created by running the code without modifying the sampler, while the one below it was created by removing the sampler and running the code. Please let me know if there is a proper way to do this.