Stable Diffusion 3.5 ControlNets
Originally posted on Huggingface.
The following control types are available:
Canny - Use a Canny edge map to guide the structure of the generated image. This is especially useful for illustrations, but works with all styles.
Depth - use a depth map, generated by DepthFM, to guide generation. Some example use cases include generating architectural renderings, or texturing 3D assets.
Blur - can be used to perform extremely high fidelity upscaling. A common use case is to tile an input image, apply the ControlNet to each tile, and merge the tiles to produce a higher resolution image. A more in-depth description of this use case is here.
For Comfy users, this extension provides support.
We recommend tiling the image at a tile size between 128 and 512.
All currently released ControlNets are compatible only with Stable Diffusion 3.5 Large (8b).
Description
FAQ
Comments (1)
In general, Blur is not Blur at all. This is an upscaler. The Workflow they offer as a sample doesn't do anything radically good. It is necessary to use Tiled Diffusion by link, then it turns out interesting.