OutfitAnyone includes try-on models and auxiliary models for virtual try-on scenarios, from quick image generation and detail refinement to partial replacement.
This document applies only to the China (Beijing) region. To use the models, you must use an API key from the China (Beijing) region.
Model overview
|
Type |
Service |
Model |
Core feature |
Scenarios |
|
Try-on model |
aitryon-plus |
Higher image definition, improved texture details, and better logo restoration. |
Slower generation speed, suitable for non-time-sensitive scenarios. |
|
|
Auxiliary model |
aitryon-parsing-v1 |
Segments clothing areas in images. |
Enables partial try-on and interactive features like product hotspots. |
Scenarios
Combine these APIs based on your needs. Common scenarios:
Scenario 1: Basic try-on
Use case: Quick result verification or high-speed image generation.
Steps:
-
Prepare inputs: Model image and clothing image.
-
Call try-on model: Call OutfitAnyone-Plus API reference with the two images.
-
Get result: The API returns the try-on image.
Scenario 2: Partial try-on
Use case: Replace part of the outfit (e.g., replace only the top, keep the original pants).
Steps:
-
Extract clothing to keep: Call OutfitAnyone-Parsing with the original model image to extract the clothing item to keep (e.g., pants).
-
Combine for try-on: Call OutfitAnyone-Plus API reference with the original model image, new top, and extracted pants image.
-
Get result: The API returns a try-on image combining the new top and original pants.
Scenario 3: Obtaining clothing coordinates
Use case: Add product labels or implement interactive features like product hotspots.
Steps:
-
Call image segmentation: Call OutfitAnyone-Parsing with any model image (real photo or OutfitAnyone-generated image).
-
Get result: The API returns clothing area coordinates (bbox) and visualization. Use these coordinates for frontend interactive features like product hotspots.