Pixel-AI uses AI to detect different color pixels on a white canvas. Training is done on synthetic images, where a single pixel of a random color is placed at a random position on a white canvas. The model learns to predict the position of the colored pixel based on the image.
No followers yet
Once you ship this you can't edit the description of the project, but you'll be able to add more devlogs and re-ship it as you add new features!
I have created the basics of training and recognition for the AI model. At the end of each training session the program outputs a matplotlib graph which shows the loss of the AI model over time.