Vibe Coding DiscoverControlNet
View on GitHubLet us control diffusion models!
Official Python implementation of ControlNet, which adds spatial conditions such as edges, depth, poses, and scribbles to diffusion models. Includes pretrained Stable Diffusion 1.5 demos and Gradio apps.
Use Cases
Guide image generation with edge mapsGenerate images from depth mapsControl composition with human posesColorize or stylize images from line artGenerate images from scribblesCondition images on semantic segmentation or normal maps
Built With
- Language
- Python
- Frameworks
- PyTorch · Gradio · Latent Diffusion
Tags
diffusion models · image generation · image conditioning · ControlNet · computer vision · image-to-image · pretrained models