Updates
With this update, CLO-SET introduces AI Visualizer (Beta), a new way to turn the 3D garments in your workspace into photorealistic images directly from the Content Page.
AI Visualizer generates realistic visuals from your 3D assets, so teams can review a style as it would look in the real world, and preview combinations with assets that are not ready yet or with external image references, supporting faster and more intuitive Sample Reviews.
This release focuses on:
AI Visualizer (Beta): Photorealistic Image Generation from the 3D Viewer
Note: AI Visualizer is released as a Beta feature and is available on paid plans only.
Details
1) AI Visualizer (Beta): Photorealistic Image Generation from the 3D Viewer
With this update, an AI Visualizer button is available in the 3D Viewer on the Content Page.
Clicking it opens the AI Visualizer panel, where the garment currently shown in the 3D Viewer is set as the Base image. From there, images can be generated in three steps:
Step 1. Add references (optional)
Alongside the Base garment, up to 8 reference assets can be added and combined into the generated image:
- From CLO-SET: browse or search your workspace and select 3D assets, such as a jacket or other garments, to style together with the Base garment.
- From Local: upload image files (max file size: 20MB), such as photos of shoes, bags, or other reference items that are not available as 3D assets yet.
.
Step 2. Select an AI engine and describe the style
- Choose an AI engine for generation: GPT Image or Gemini.
- Describe the image you want in the prompt field, or start from the provided style presets such as Street Fashion, Cinematic Scene, Moody, Technicolor, and Runway.
Additional options give more control over the result:
- Preserve avatar face: keep the avatar’s face consistent in the generated image.
- Aspect ratio: select the output ratio, such as 3:4.
Step 3. Generate and save
Click Generate to create the image. Results appear in the Generated panel, where they can be selected, saved, and downloaded.
Note: Images generated on a content are visible to the members of the group that content belongs to.
Step 4. Refine the result (optional)
Generated images can be refined further instead of starting over. Check Use current image as base to generate from the image you just created, and set AI Transformation Strength to control how much the AI changes it:
- Default: a balanced level of transformation.
- Preserve Original Structure: keeps the garment and composition close to the source, for finer touch-ups.
- Allow Creative Transformation: gives the AI more freedom for a stronger restyle.
This allows a result to be developed step by step, keeping what already works and adjusting only the details.
This enhancement allows teams to review styles and styling combinations as photorealistic images before a physical sample exists, and to explore looks that would otherwise require a photoshoot.
AI Visualizer works alongside the existing Render and mix&match features, and each answers a different question: Render produces images that stay faithful to the 3D garment exactly as simulated, mix&match styles the 3D assets already in your workspace, and AI Visualizer explores how a style could look on a model, in a scene, and together with items that do not exist as 3D assets yet.