Product Studio

Virtual Try-on

Generate visualization images by placing clothing or accessory assets onto a person image for product presentation and ad testing.

Route: /virtual-try-onOpen feature

Overview

What it does: Generate visualization images by placing clothing or accessory assets onto a person image for product presentation and ad testing.

Where to find it: Navigate to /virtual-try-on from the Dashboard sidebar.

Step-by-step guide

  1. Upload one clear person/model image as the base.
  2. Upload one or more product assets (up to 5 files) you want to apply.
  3. Select the target aspect ratio for your destination channel.
  4. Generate, inspect fit quality, and rerun with cleaner source files if needed.
  5. Download approved outputs for product pages, ads, or social use.

User Inputs (What users enter)

  • Person Image/Model (Base photo of a person)
  • Assets (Up to 5 clothing/accessory images)
  • Aspect Ratio (Target output size)

UI Breakdown

Primary Input

Provide clear business context and the exact goal for this feature.

Options Panel

Configure available settings to match brand style and campaign intent.

Generate and Review

Generate outputs, compare versions, and keep the strongest result.

How it works

  1. Step 1: Provide a base photo of a person.
  2. Step 2: Upload images of the clothing or accessories to be tested.
  3. Step 3: AI intelligently wraps and maps the assets onto the person's body.
  4. Step 4: Generate a realistic visualization for product pages or ads.

Output

You get ready-to-use output generated from your inputs and selected options, which you can refine and publish in your workflow.

Example output

Sample result for Virtual Try-on appears here after generation and can be reused in campaigns.

Screenshots / GIFs

Add screenshot: feature main screen
Add GIF: step-by-step flow

Tips / Notes

  • Use front-facing or clear-angle poses to improve asset mapping quality.
  • Avoid heavily occluded clothing shots; full-item visibility gives better results.
  • If alignment looks wrong, retry with higher-resolution, better-lit source images.