Models & ArchitectureUpdated 2026-10-24
ControlNet
An add-on that gives diffusion models structural control from pose, edges, depth and more.
#ControlNet#可控生成#图像
In plain language
ControlNet makes generation follow the structure of a reference image instead of text alone. Common controls include Canny edges, OpenPose, depth and scribbles. Stronger control means less creative freedom.
Example
Lock a pose with OpenPose, then change clothing and scene via prompt.
Recommended tools
Related terms
Models & ArchitectureFeatured
Diffusion Model
A generative model that turns random noise into images, video or audio through iterative denoising.
3 related terms
Models & Architecture
Latent Space
A compressed high-dimensional representation space where diffusion often operates.
3 related terms
Models & Architecture
Low-Rank Adaptation
Freezing the base model and training a small low-rank adapter for a new task.
3 related terms
Applications
Agentic Workflow
Breaking a task into steps planned by AI, executed by tools and reviewed by humans.
3 related terms