elixir-image / image_detection

Image object detection for Elixir using Axon and the Yolo V8 model

Image Detection

Image Detection is an Image-based object detection library based upon the YOLO V8 ML model.

The documentation can be found at https://hexdocs.pm/image_detection.

Installation

Image Detection can be installed by adding image_detection to your list of dependencies in mix.exs:

def deps do
  [
    {:image_detection_, "~> 0.1.0"}
  ]
end

Configuration

THIS IS AN EXPERIMENTAL LIBRARY. Please do not use in production. Testing is not yet complete.

The code is an adaptation of the livecoding demo by Hans Elias (@hansihe) at his talk at the Elixir Warsaw meetup in March 2023.

Accessing the Yolo V8 model

The YOLO v8 model is GPL 3.0 licensed and must be built separately. Python 3.10 is required (apparently it won't work with 3.11).

% pip3.10 install ultralytics
% pip3.10 install onnx
% yolo export model=yolov8n.pt format=onnx imgsz=640

The "n" model is the smallest - we can maybe tolerate a larger one. And we need to find way to host the model or download the .onnx from somewhere.

References

• The original talk by @hansihe is at https://www.youtube.com/watch?v=OsxGB6MbA8o&t=1s
• The original models are stored at https://github.com/ultralytics/assets/releases
• Learning: https://learnopencv.com/ultralytics-yolov8/
• Exploration: https://medium.com/mlearning-ai/yolo-v8-the-real-state-of-the-art-eda6c86a1b90