This toolset provides a professional suite of Houdini Digital Assets (HDAs) for integrating Stable Diffusion image generation and processing directly within the Houdini environment. It targets 3D artists and technical directors seeking to leverage AI image synthesis for asset creation, animation, and complex visual effects pipelines, offering seamless integration with popular Stable Diffusion interfaces like Automatic1111 and ControlNet.

How It Works

The toolset leverages Houdini's Procedural Dependency Graph (PDG) to orchestrate Stable Diffusion tasks, enabling complex, automated workflows. It acts as an interface to Automatic1111's API, allowing users to control text-to-image, image-to-image, ControlNet preprocessors and models, LoRA/TI/Hypernetwork usage, and upscaling directly from Houdini nodes. This procedural approach facilitates animation of prompts, bulk processing, and integration with existing Houdini scene data.

Quick Start & Requirements

• Installation: Install Automatic1111 web UI with the --api flag. Install HDAs from /hda/ and ensure Python/Presets folders are alongside top_stable_diffusion.hda. Restart Houdini.
• Prerequisites: Houdini 19.5, 20, or 20.5 with Python 3.9. Automatic1111 web UI (tested with version as of July 2024). Optional: Kohya_ss web UI for training.
• Resources: Requires a running Automatic1111 instance.
• Docs: Troubleshooting Checklist, Wiki
• Demos: Video Tutorials, Additional Demos

Highlighted Details

• Full integration with Automatic1111 features: TI, Loras, Hypernetworks, face restoration, tiling, hires fix, SD XL.
• Supports the entire ControlNet 1.1 suite, including all preprocessors and models.
• Enables LoRA training dataset preparation and LoRA training directly within Houdini via a Kohya_ss connector.
• PDG integration allows for automated, scalable generation and processing tasks, including prompt animation and bulk operations.

Maintenance & Community

• Developed by Stanislav Demchenko (stassius) for EVR Systems.
• Active Discord community available: Discord

Licensing & Compatibility

• License not explicitly stated in the README. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The toolset requires a separate, running instance of Automatic1111 and is dependent on its API compatibility. Compatibility with future versions of Automatic1111 or ControlNet is not guaranteed and may require updates to the toolset. The README notes that users should update both ControlNet and Stable Houdini to avoid compatibility errors.