Update AGPL Licensed Fork

[samphonic]

Hi there, apologies for creating an issue in the models repo for a question about your Frigate fork, but I wasn't sure where else to contact you.

I'm interested in testing an Open Images trained edgetpu model in Frigate and your models are the only I've come across. However I'm also pretty attached to the new Frigate UI in 0.14.x. I briefly tried my hand at merging the latest dev changes from the official Frigate repository with your AGPL repo, but don't understand enough of the changes you made to intelligently merge them.

How difficult would it be to merge the changes from the original repo into yours? If they've diverged a bunch or it's a whole thing, then that's understandable-- just figured it was worth asking.

(FWIW, I don't really understand how only adding support for YOLOv8 models to Frigate requires it be AGPL licensed. Surely non-AGPL projects can use a YOLOv8 model if they're not distributing/modifying it, no?)

[samphonic]

Actually, a related follow-up question (sorry if this is an obvious thing, I'm still a bit green in ML):

It's possible for me to drop-in a e.g. spaghettinet model and tell Frigate to use it, but when I try the same with one of your yolov8 models there's a UINT/INT error (and maybe more). Without worrying about "optimizing" Frigate's use of the model, is there a minimal way to just chuck it in there and see how it goes?