None segmentation version

[zan47der]

Hi author!
Would you mind adding a version without segmentation? Only detection bounding boxes are involved.

[TimoK93]

Hello Alexander,

thanks for your suggestion. Do you have a specific data format in mind?

Cheers,
Timo

[zan47der]

Hello Alexander,

thanks for your suggestion. Do you have a specific data format in mind?

Cheers, Timo

Hello Timo,
Thanks for your contribution about CHOTA metric, I would appreciate it if you can consider incorporating the calculation of MOT bbox format (frame, id, bb_left, bb_top, bb_width, bb_height) into your code.

[TimoK93]

Hi,

is there a specific data format in which you are interested? There are formats like in the e.g. the MOTChallenge (CTMC), or in other benchmarks like KITTI.

I think that the change is not trivial as the code base is designed to satisfy the CTC requirements. I will overthink it when I have the time for it (which is unfortunately not too much :( ). Meanwhile, feel free to contribute and open a PR if you see a simple workaround that satisfies your needs!

Best,
Timo

[zan47der]

Hi author,
The CTMC format is needed, thanks.

[TimoK93]

Hello Alexander,

I created a development branch for CTMC evaluation here https://github.com/CellTrackingChallenge/py-ctcmetrics/tree/CTMC

Could you run the code on your results and report how it behaves?

[TimoK93]

@alexandermmcdxviii
A description on how to use it is in the Readme

[zan47der]

Hello Alexander,

I created a development branch for CTMC evaluation here https://github.com/CellTrackingChallenge/py-ctcmetrics/tree/CTMC

Could you run the code on your results and report how it behaves?

Thank you! I'll see what I can do.

[TimoK93]

Is the code doing well?