search

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
3.56k stars 598 forks source link

Warning: The model doesn't have classes #475

Open ghost opened 2 years ago

ghost commented 2 years ago

I get this warning (Warning: The model doesn't have classes) for every tracktor and deepsort model, when I run the following:

python demo/demo_mot_vis.py configs/mot/deepsort/sort_faster-rcnn_fpn_4e_mot17-public-half.py --input demo/demo.mp4 --output mot.mp4 --show --backend plt

Am I doing something wrong?

JingweiZhang12 commented 2 years ago

It's normal. The warning may be caused by the detector of MOT being trained on an earlier version of mmdet.

ghost commented 2 years ago

Thank you for the quick reply. When testing mmtracking, no labels are assigned to the objects. Is there any way to change this? Do I have to retrain the mmdet model?

Am Di., 29. März 2022 um 14:42 Uhr schrieb Jingwei Zhang < @.***>:

It's normal. The warning may be caused by the detector of MOT being trained on an earlier version of mmdet.

— Reply to this email directly, view it on GitHub https://github.com/open-mmlab/mmtracking/issues/475#issuecomment-1081823494, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADMLJTPU4OABLPIMR2YIEWLVCL3CHANCNFSM5R5TKG4Q . You are receiving this because you authored the thread.Message ID: @.***>

JingweiZhang12 commented 2 years ago

Did you test MOT model? Please post the details about your problem.

ghost commented 2 years ago

Yes, I did.

Jingwei Zhang @.***> schrieb am Mi., 30. März 2022, 14:49:

Did you test MOT model?

— Reply to this email directly, view it on GitHub https://github.com/open-mmlab/mmtracking/issues/475#issuecomment-1083099376, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADMLJTLDVHZZ3755XZF4TEDVCREWXANCNFSM5R5TKG4Q . You are receiving this because you authored the thread.Message ID: @.***>

jinweiisgreat commented 1 year ago

you need to add a checkpoint file behind the config file. like this: python demo/demo_vid.py \ configs/vid/selsa/selsa_faster_rcnn_r50_dc5_1x_imagenetvid.py \ --checkpoint selsa_faster_rcnn_r50_dc5_1x_imagenetvid_20201227_204835-2f5a4952.pth \ --input demo/demo.mp4 \ --output vid.mp4