Virtual try- ons
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass What I Built A virtual try-on web app. You upload a photo of yourself and a photo of a garment, and it shows you how the outfit might look on you. The results are image, size or fit suggestion How it gets people off the screen: Shopping and outfit-picking usually mean an hour of scrolling, ordering multiple…
This submission is for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass. It presents a virtual try-on web app that allows users to upload a photo of themselves and a photo of a garment. The app then displays how the outfit might look on the individual, providing an image, size, or fit suggestion. By reducing the need for scrolling, multiple size orders, and returns, the virtual try-on process takes only a minute, allowing users to close their laptop and step outside, be it to the market, campus, or a trail.
The virtual try-on app is aimed at college students and online shoppers who wish to minimize returns and reduce screen time. Initially, the app was built using a simple vibe code, which functioned and looked good but lacked real AI for try-on capabilities. However, subsequent improvements focused on integrating an actual open model.
The core of the app relies on open-weight AI, specifically Gemma, an open-weight model developed by Google. This model analyzes the garment or photo, generating outfit suggestions and providing detailed fit descriptions. The remaining components of the application's stack, apart from the open-source AI component, are not open-source.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
