Could mmtrack eval the ImageNetVID by slow/medium/fast group as in FGFA?

open-mmlab / mmtracking

OpenMMLab Video Perception Toolbox. It supports Video Object Detection (VID), Multiple Object Tracking (MOT), Single Object Tracking (SOT), Video Instance Segmentation (VIS) with a unified framework.

https://mmtracking.readthedocs.io/en/latest/

Apache License 2.0

3.52k stars 591 forks source link

Could mmtrack eval the ImageNetVID by slow/medium/fast group as in FGFA? #331

Open Memz99 opened 2 years ago

Memz99 commented 2 years ago

or have there already any api could do that?

GT9505 commented 2 years ago

We don't support eval mAP by slow/medium/fast group for now.

Would you like to support the feature? The FGFA provides the API in here, and here is an example.