🔮The Codex
ControlNet
A technique for guiding AI image generation with structural inputs like poses or edges.
📖 Apprentice Explanation
ControlNet lets you control AI image generation more precisely. Instead of just describing what you want, you can provide a sketch, pose, or edge map to guide the exact composition of the generated image.
🧙 Archmage Notes
ControlNet adds conditional control to diffusion models by training adapter networks on paired data (e.g., Canny edges, depth maps, poses). Supports multi-condition generation. Widely used in Stable Diffusion workflows via ComfyUI and Automatic1111.