Talk:Virtual dressing room

Untitled

Alot of information outdated on virtual fitting room technology Galadedrid Damodred (talk) 14:44, 23 May 2013 (UTC)

Wiki Education assignment: Research Process and Methodology - RPM SP 2022 - MASY1-GC 1260 201 Thu

This article was the subject of a Wiki Education Foundation–supported course assignment, between 27 February 2022 and 5 May 2022. Further details are available on the course page. Student editor(s): Nuts666 (article contribs).

Edit request: section on generative AI try-on

I'm the founder of Genlook, a vendor of AI virtual try-on software, so per WP:COI I'm proposing this here rather than editing the article directly. The article covers size recommendation, body scanning, 3D, AR and photo-based approaches, but nothing on generative AI (diffusion-model) try-on, which reached production from 2023 (Google Shopping) and shoppers' own photos in 2025 (Google, Zara, later ASOS). Proposed section below, to be placed after "Augmented reality". Sources are almost all independent (CVPR papers, CNBC, TechCrunch, Axios, Retail Dive); the sentence naming my company is cited to CNBC's April 2026 coverage, and one supporting citation on generation speed points to a data report we published — happy to have that one dropped if reviewers consider it too promotional.

This LLM-generated text has been collapsed and should be excluded from assessments of consensus.

The following discussion has been closed. Please do not modify it.

Generative AI

Image-based virtual try-on, in which software synthesises a picture of a person wearing a given garment from separate photos of the person and the garment, was an active research topic from the late 2010s, first with encoder-decoder networks such as VITON (2018), later with generative adversarial networks such as VITON-HD (2021). Diffusion model-based systems such as TryOnDiffusion followed in 2023, and Google launched generative try-on in Google Shopping the same year, initially rendering garments on real catalogue models rather than on the shopper's own photo.

For clothing, generative try-on on the shopper's own photo first reached production in 2025. Google launched "try it on" in the United States in May 2025, letting shoppers see garments on a full-length photo of themselves, and later extended it to more countries, shoes, and a selfie-based flow powered by its Gemini image generation model, Gemini 2.5 Flash Image ("Nano Banana"); its standalone app Doppl (2025) was folded into Search the following year. Retailers have adopted the technology partly to reduce product returns, which the National Retail Federation estimated at 15.8% of United States retail sales in 2025, or $849.9 billion. Zara added try-on to its app in December 2025, reporting more than seven million sessions across 43 markets by March 2026, and ASOS introduced a hybrid feature in 2026 in which shoppers upload their own photo or use an AI-generated model matching their size. Independent online stores can add generative try-on through e-commerce platform plug-ins, such as the startup Genlook's try-on app for Shopify; the shopper uploads a photo on a product page and receives a generated image of themselves wearing the product, typically in around ten seconds.

References

Thibaultmthh (talk) 13:56, 20 July 2026 (UTC) Thibaultmthh (talk) 13:56, 20 July 2026 (UTC)

Reply 1-AUG-2026

  Edit request declined  

• The proposed text appears to have been generated by an LLM.

Regards,  Spintendo  16:45, 1 August 2026 (UTC)