ghost
commented
4 years ago Btw, this project would be mainly useful for robotics, VR, and AR, so these are really some real world technologies, but I still think it would be useful, even though this is a slightly more intermediate topic, just because you never know what people might want to do with computer vision.
In this challenge, if you choose to accept it:
You must create a processing sketch with realtime webcam tracking using only landmarks in the image.
You should be able to get the position and the rotation of the webcam.
It should work with a standard monocular webcam.
This is more of a challenge in getting this sort of thing running in processing. I don't expect you to write your own fully fledged SLAM implementation, but it would be awesome if you did, of course!
That said, maybe the second part to this challenge could be to make it production ready. AKA, using only code that is commercially licensable, and more importantly, making it work outside also, and having it ignore moving objects such as people and cars.
I think you can do this. Give it a shot! :)