Hey, this is great :) I attempted to do something similar a while back https://github.com/aadv1k/deimos basically trying to build many of OpenCV's functions from scratch in C from first principles, though I was using stb for handling the images. I ended up putting the project on hold, primarily because I lost interest in computer vision at the time.

For a while I went deep into OCR, and built a rather rudimentary stroke width transform (https://github.com/aadv1k/swt.h) but again, the results were very hit or miss, likely because I never took the time to understand the logic behind why these functions would work.

1) Optimise many of the functions (a lot of room to use GPU, multi-threading and what not!). 2) Add new functions and improve the existing edge detection ones

I would love to know of a good resource for computer vision, the various algorithms, optimisation techniques etc. Thanks for sharing this project! Cheers

Since this is intended for embedded systems, supporting ARM DSP extensions would be beneficial. Correct me if I'm wrong, but I have not seen any compiler generate those instructions, other than through the use of intrinsics.

I would add the support if I needed the library, but I don't, at least not yet.

I basically did the same thing a few weeks ago! =) => https://flatcv.ad-si.com

It will be interesting to see what you did differently!

Did you see the "By The Power of Grayscale" submission and go digging?

cf. https://en.wikipedia.org/wiki/Castle_Grayskull

I’d like to take this moment to say that the recent She Ra revival series on Netflix by ND Stevenson (the creator of Nimona) is pretty good, go watch it.

On this tip I always found opencv to work way way faster just dropping the color depth of whatever image as such

Missed opportunity to one-up the He-Man joke from earlier :p

related: https://news.ycombinator.com/item?id=45771151