Depth map clip-art for Stable Diffusion pose control

What it does This extension adds a depth-map library interface to the Automatic1111 WebUI, letting you select and composite pre-made silhouettes without leaving the browser. It feeds the resulting depth map straight into ControlNet to guide generation. You can also permanently add your own assets by dropping them into maps/<category> folders.

The interesting bit It treats depth maps like clip art: instead of extracting depth from a photo, you assemble a scene from reusable 2.5D assets and let the model infer the rest. The README’s only operational warning is to leave ControlNet’s preprocessor empty, which reveals the tool is essentially a raw-depth bypass lane.

Key highlights

• Provides a visual depth-map picker integrated directly into the Automatic1111 WebUI
• Supports custom libraries by adding images to maps/<category> folders; the repo includes a shapes example
• Acts as a raw-depth source for ControlNet, requiring no preprocessor
• JavaScript plugin that runs entirely inside the existing WebUI environment

Caveats

• The README mentions a “poser” but never explains what it does, what controls it offers, or what maps ship with the library beyond a single screenshot.
• A typo in the folder path (examle) and the very sparse docs suggest the project is lightly maintained.

Verdict Handy if you generate character or object poses in Stable Diffusion and want a reusable depth shortcut without preprocessing photos. Skip it if you need detailed documentation or advanced depth editing; this is a simple helper, not a full toolkit.