Research
Virtual try-on benchmark: access, inputs, and use cases
A source-backed comparison of what five virtual try-on products document—not a subjective image-quality ranking.
Checked July 22, 2026 · By TryDrobe Editorial Team
What this benchmark measures
“Virtual try-on” describes several different product shapes: a consumer app that accepts any garment image, a shopping feature tied to eligible listings, a live camera effect, or an API for retailers. Comparing them as if they perform the same job hides the most important distinction: how a person accesses the product and what inputs it accepts.
This benchmark records four facts available from first-party materials: primary access, person input, clothing input, and intended use case. It does not score realism, fit accuracy, speed, privacy, or current geographic availability. Those claims would require a controlled test or additional evidence beyond the documentation review used here.
Comparison table
| Product | Primary access | Person input | Clothing input | Documented use case |
|---|---|---|---|---|
| TryDrobeSource | TryDrobe web app and iOS app | User photo | Garment or outfit image supplied by the user | Consumer virtual preview, digital closet, and outfit planning |
| Google Shopping try-onSource | Eligible apparel listings in Google Shopping | Full-length photo; Google also documents model selection | Eligible Shopping product listing | Consumer shopping across supported apparel categories |
| Google DopplSource | Experimental Google Labs mobile app; official launch page describes U.S. availability | User photo | Outfit image or screenshot supplied by the user | Experimental consumer outfit visualization and AI video |
| Amazon Virtual Try-On for ShoesSource | Amazon shopping experience documented at launch in the iOS app | Live phone camera view of the user’s feet | Supported shoe listing | Consumer AR visualization for shoes |
| Perfect Corp AI Fashion Try-OnSource | Business product and API | Model, avatar, or customer image | Garment image | Retail and brand integration |
Amazon's cited page documents its 2022 launch. Google Doppl's cited page describes the experimental app's U.S. launch. Check each product directly for current access.
What the access model tells you
- Independent consumer tool: useful when the garment source is not tied to one retailer.
- Shopping integration: useful when the item already appears in an eligible commerce catalog.
- Camera AR: a different interaction from generating a still image from uploaded photos.
- Business API: intended for a retailer or brand to add try-on to its own experience.
What this data cannot tell you
- Which product makes the most realistic image
- Whether a garment will physically fit
- How each product handles every image, body, or garment type
- Whether a feature is available in every country or account
Methodology
- Include products with a first-party page that clearly documents a fashion try-on workflow.
- Record only statements supported by that product's official page or documentation.
- Normalize the statements into the same four fields without assigning a score.
- Flag dated launches and experimental availability instead of treating them as permanent.
TryDrobe is included and is our product. Its row is based on the current TryDrobe product experience. The limitation applies equally to every row: documentation can describe a workflow, but it cannot substitute for a controlled image-quality test.
First-party sources
Continue the research
Use the best virtual try-on apps guide for a consumer-oriented shortlist, or the virtual fitting room app page to understand TryDrobe's product workflow.