What You’ll Learn?

This workshop is designed for beginners with no prior knowledge of ComfyUI. The aim is to produce precise results of an existing site image or a project image in the form of images, videos, and 3D models, with most of the work carried out locally to ensure security and control.


Here’s what you’ll specifically design:

We begin with an introduction to the fundamentals—checkpoint models, latent space, VAEs, and prompt structures—so that participants gain a deeper understanding of how ComfyUI operates beneath the interface. From there, the workflow unfolds in three stages. The first is the image stage, where participants will use ControlNet modules such as Depth and Canny to guide the AI in generating design iterations that respond to time, season, and material conditions. Flux tools will then be applied for style transfer, inpainting, and upscaling to refine the outputs into coherent, detailed visuals.

Second is the video stage, where still images will be converted into moving sequences using WAN 2.2. Participants will learn prompt structures, explore model variations, and adjust settings such as frame rate and camera motion to generate videos that add more detail to their designs. Finally comes the 3D stage, where images will be prepared with Gemini 2.5 Flash Image (Nano Banana) for 3D interpretation, then converted through Hunyuan 3D or Sparc 3D. The resulting models will be analyzed, refined, and cleaned in Rhino3D (or another preferred method) to create fabrication-ready geometry. By the end of the workshop, participants will have completed the full cycle—text-to-image, image-to-image, image-to-video, and image-to-3D model—transforming their designs into human–AI co-created outputs by leveraging AI’s latent capabilities.

By the end, participants will have worked through the full cycle—text-to-image, image-to-image, image-to-video, and image-to-3D model—ultimately transforming their designs into human–AI co-created outputs by leveraging AI’s latent capabilities.