Question: I am looking to start some research in the image processing domain, for which I am currently trying to find a problem. I would love to know what other folks are working on!

@warren

For consideration: - Image processing/computer vision courses that discuss lots of different research topics:
https://github.com/MargaretAN9/Peggy/blob/master/computer%20vision%20course.md -Applying AI to an environmental sorting problems (microplastics/ algea/etc ) using techniques such as: https://medium.com/@timanglade/how-hbos-silicon-valley-built-not-hotdog-with-mobile-tensorflow-keras-react-native-ef03260747f3 or https://cloud.google.com/vision/

Thanks a lot @MaggPi these resources are really nice!! ❤️

Perhaps @amirberagain might also have some suggestions!

Thanks for pinging me. The list @maggPi posted looks great and I didn't know most of them. However, I personally like to get some experience before diving in the theory of it all (which is necessary, don't skip it!). One resource I really liked is the generic OpenCV tutorials, very basic and provides a good background of basic tools needed. An additional great resource is Adrian Rosebrock's blog - PyImageSearch where he recently started diving into DL as well. I'd recommend you take time to study CV before DL, this background is valuable.

A couple ideas - what about classification (potentially as part of imagesequencer?) -- lots of interest in this for aerial photography but generally doesnt run in the browser (though see this prototype: #clashifier) for quantifying land cover types.

Second, interest point finding and matching between image pairs -- there's an issue open here and it's a very big domain:

https://github.com/publiclab/image-sequencer/issues/300