Stable Diffusion and Friends - High-Resolution Image Synthesis via Two-Stage Generative Models

This course aims to teach learners about high-resolution image synthesis through two-stage generative models. The learning outcomes include understanding the history of generative image models, such as GANs, Transformers, and latent Diffusion models. The course covers individual skills like working with QCVAE architecture, Vision Transformers, VQGAN, and Stable Diffusion. The teaching method involves a guided tour by a research scientist, focusing on generative deep learning models. The intended audience for this course includes individuals interested in generative image models and text-to-image systems.