Procedural Fragment Shader Generation Using Classic Machine Learning

[k0T0z]

As mentioned in A Tool for the Procedural Generation of Shaders using Interactive Evolutionary Algorithms
paper, this project is divided into two major steps:

1. Implement the visual shader graph functionalities.
2. Apply a proper Genetic Algorithm to find the best shader for a pre-specified image.

The contributor may also read Noise Modeler: An Interactive Editor and Library for Procedural Terrains via Continuous Generation and Compilation of GPU Shaders paper for a better understanding of the graph editor architecture.

Mentors: Robert, Josh, Greg
Difficulty: Medium - Hard
Expected size: 350h
Skills required: C++ fundamentals, Classic Machine Learning, Qt
Skills preferred: Knowledge of Genetic Programming is preferred as well as Deep Learning theory