ControlNet is a neural network architechture that can be used to control pretrained large diffusion models to support additional input conditions. The purpose of ControlNet HED is to use the soft HED Boundary to preserve many details in input images, making this model suitable for recoloring and stylizing.
The model is trained with HED boundary detection to obtain 3M edge-image-caption pairs.
ControlNet HED can be utilized in various applications, such as:
For more detailed instructions, refer to the API documentation and resources available on Github.