The Yoga of Image Generation with Stable Diffusion and ComfyUI

Explore controlled image generation using Stable Diffusion and ComfyUI in this hands-on conference talk that demonstrates how generative AI can create accurate, diverse, and inclusive images of yoga poses. Learn to harness ComfyUI's modular workflows, embeddings, inpainting, ControlNets, LoRA, and FaceID for precise style and pose control when stock photos or internet-sourced images fall short. Master both text-to-image and hybrid image-to-image workflows while discovering solutions to key challenges including pose consistency and identity preservation. Gain practical insights from the development of YogĀrkana, a dedicated yoga website, through live demonstrations and actionable tips. Develop the skills to customize Stable Diffusion workflows and run them locally to achieve your creative, commercial, or inclusive design objectives with precision and ease.