Description

ControlNet V1.1 is a Hugging Face Space that lets you upload a picture and use a guide such as an edge map, sketch, depth map, or pose to create a new image that follows that guide.

The visible page indicates guide types are selected from tabs such as Canny, Scribble, depth, and pose.

How we innovate

Its distinctive feature is direct control over image generation through visual guides like edges, sketches, depth maps, and poses rather than text alone.

Use Case / Scenario

1. Guided image generation

Use an input image together with a guide to produce a new image that follows the chosen structure.

2. Edge, sketch, depth, or pose control

Select a guide type from the available tabs, such as Canny, Scribble, depth, or pose, depending on the kind of structure you want to preserve.

3. Image transformation experiments

Try different guide inputs to explore how the model changes an image while keeping it aligned with the provided reference.

Visit Website