List of computer vision pretrained models

Thanks for sharing. Since it's your own repository I'd recommend to append a "Show HN:" at the beginning.

I've also learned of Netron to visualize a network's architecture. That's neat way to get a grasp of a new model (here a screenshot: https://raw.githubusercontent.com/lutzroeder/netron/master/....)

A list is useful, but would be even more so if there were some indication of the domains in which a model may be useful.

If the model description included the problem they were trying to solve, that would let me judge how closely it mirrored my own requirements. If I saw "We wanted to build something to recognize flags in photographs" then I'd know that's something that might help me build something to recognize company logos (for example).

Well, see:

- https://paperswithcode.com/ if you care for state of the art results

- https://pytorch.org/hub/ if you care for direct import into PyTorch code

A list of models, without performance metrics, is not nearly as useful.

Google's ambitious BiT (Big Transfer) model is missing from the list:

https://ai.googleblog.com/2020/05/open-sourcing-bit-explorin...

You can also check https://github.com/Cadene/pretrained-models.pytorch.

Models are directly loadable into pytorch.