using yolo for assisted labeling

[williamhoole]

Actions before raising this issue

• I searched the existing issues and did not find anything similar.
• I read/searched the docs

Is your feature request related to a problem? Please describe.

I am not that experienced with CVAT but i see you are able to do assisted labeling with a pre-trained model. However the process of getting this to work is pretty complex and i havent gotten it to work. I would like to use either a yolov5 or yolov8 model from ultralytics to assist during my annotation process. I have the weights locally on my computer and wold like to import it to cvat. Is it possible to do this online? or will i have to download cvat with all its docker containers and get that to running... followed by the nucli server etc. could someone assist on how to do this in a more simpler way? Since Yolo is probably the most common object detection framework.

Describe the solution you'd like

No response

Describe alternatives you've considered

No response

Additional context

No response

[davy-blavette]

It seems to me that CVAT doesn't offer Yolov5 or Yolov8 for reasons of license incompatibility, but it is possible to use them locally if you set the serveless correctly.

[O-J1]

It seems to me that CVAT doesn't offer Yolov5 or Yolov8 for reasons of license incompatibility, but it is possible to use them locally if you set the serveless correctly.

Do you happen to have or know of any guidance on this?