Preview an outfit from
saved product photos.

What this workflow is

It is a short decision workflow for pieces you have already found online or photographed yourself. The images can stay in your phone's photo library until you choose to upload them for a preview. Private Try-On does not automatically read the rest of your photo library or turn it into a digital closet.

The output is one generated style direction. It can help answer “does this combination make visual sense?” It cannot answer “which size fits?” or prove that the delivered products will look identical.

1. Save only the pieces you are deciding between

Create a temporary album on your phone and save one image for each role: person reference, top, bottoms and shoes. Add an outer layer or accessory only when that option is enabled and actually matters to the decision.

A smaller set is easier to compare and easier to delete afterwards. This is different from building a permanent digital wardrobe: there is no need to label every garment, track wear history or upload clothes unrelated to the current outfit.

2. Choose images the model can read

• Use one clear, front-facing full-body person photo with the head and feet visible.
• Prefer product images where the garment is unobstructed and fills most of the frame.
• Crop away prices, store buttons, sale badges and unrelated text.
• Avoid collages when separate top, bottoms and shoe inputs are available.
• Use only photos you own or have permission to process.

Clean inputs do not guarantee accuracy, but they reduce ambiguity. A product image with three garments, a model and a large promotion banner gives the generator several conflicting signals.

Try a manual collage before generating

If you only need to check whether the colors and shapes feel coherent, start with a free collage or mood board. Remove the image backgrounds when practical, place the actual top, bottoms and shoes next to one another, and look for one item that feels visually disconnected.

A collage is fast and keeps the source products easy to inspect. Its limit is equally important: flat product images do not show how the combination reads on a person, and they still cannot tell you whether a size will fit. Move to an on-person style preview only when that additional visual context is useful.

3. Upload each image in its matching role

Add the person reference first, then the top, bottoms and shoes. The separate roles tell the workflow which images belong to the same outfit without pretending that the app has scanned your whole closet.

Open the fitting room and add your selected pieces →

4. Read the result as a direction, not proof

First look at the whole: color balance, visual weight and whether one item feels disconnected. Then compare the generated image with every source image. Check the neckline, print, trouser shape, shoe silhouette and any logo or trim.

If an important detail changed, the preview is not reliable evidence for that detail. You can simplify the input or try another generation, but another attempt uses the applicable allowance and still cannot guarantee fidelity.

What happens to the photos?

Private Try-On does not publish uploaded images or place them in a public gallery. The images are still sent to an external AI provider to generate the result, and provider retention or model-training rules may apply. Review the current privacy and processing explanation before uploading a personal photo.

Questions people ask