Painting with GANs - Challenges and Technicalities of Neural Style Transfer

Explore the fascinating world of Generative Adversarial Networks (GANs) and Neural Style Transfer (NST) in this 44-minute talk from EuroPython 2020. Delve into the challenges and technicalities of NST, learning how to teach machines to paint images and generate artistic artifacts. Begin with a brief introduction to GANs before diving deep into Style Transfer concepts, NST networks, and crucial loss functions including Content, Style, and Total Variation. Witness a code walkthrough with result analysis, and discover the real-world applications and challenges of this technology in gaming, fashion, design, and mobile applications. Gain valuable insights into this commercially applicable technique that goes beyond generating fake content or increasing image resolution.