Cocoa Xu

A UwU Software Engineer

Cocoa is a software engineer, and there was a time when they needed to do some image-processing tasks in Elixir, but there were no Elixir libraries that could satisfy their demands. So Cocoa developed this Evision library to address image-processing tasks in Elixir using OpenCV.

Previously, there were mainly two ways to use OpenCV in Elixir: creating a helper process (written in C++, Python or other languages) and communicating through a pipe, or manually porting OpenCV functions as NIFs, native-implemented functions so that Erlang and Elixir can load and use them directly.

These two approaches have different pros and cons, and this library focuses on the latter and tries to automate the whole porting process. later received 1-year funding from the Erlang Ecosystem Foundation.

Evision One Year - An Evolving Elixir Library for Computer Vision Tasks

One year into the development of the Evision library, evision is actively integrating itself with Nx, Kino and Livebook to create an easy-to-learn computer vision library with visualised results and several out-of-the-box Kino widgets. Meanwhile, it now supports not only OpenCV core modules but also extension modules from the opencv_contrib repo. That includes these CUDA modules that enable processing images or inferencing neural networks on an NVIDIA GPU.

This session will cover 3 main points, 1) all out-of-the-box Kino widgets (as of 1st Jan 2023, there are 5 of them, one of which is a collection of 6 different neural network tasks. New widgets will also be included if the session time allows.) 2) how to use Evision with Nx to create a machine learning + computer vision pipeline. I’ll do a live demo that shows how to solve a sudoku game (the puzzle will be captured using camera input) on Livebook. 3) Show and compare the differences when processing images using CPU, OpenCL and CUDA.