NetGen

Corey, CEO of Spellbrush, explains how the company uses deep learning to help artists create character art and variations faster. The talk introduces generative adversarial networks (GANs), where a generator learns to create images and a discriminator learns to distinguish generated images from real art. Corey describes how Spellbrush trains its models on anime illustrations, adjusts generation to improve representation, and explores animation tools. He also outlines the company’s in-house training cluster and software pipeline, and previews its plan to build an AI-illustrated game.