Wearwise

Inspiration

Online fashion shopping often involves guesswork. Wearwise was built to help users see how clothing may look on them before purchasing, while receiving recommendations that suit their occasion, preferences, and skin undertone.

What We Built

Wearwise is an AI-powered virtual try-on platform. Users upload a photo, choose an occasion or upload a specific garment, and receive photorealistic try-on results powered by the YouCam Cloth Virtual Try-On API.

The platform also analyses skin undertones, recommends relevant garments from a curated catalog, and provides an AI stylist chat. Users can refine their request naturally—for example, “Show me a navy outfit for a wedding”—and receive updated recommendations and try-on renders.

How We Built It

Wearwise uses React, Vite, Tailwind CSS, Node.js, Express, and PostgreSQL. We integrated YouCam's Virtual Try-On and Skin Analysis APIs for rendering and personalization. Google Gemini powers the AI stylist, while our recommendation engine ranks garments by occasion, gender, colour preferences, and undertone compatibility.

Challenges and Learnings

The main challenge was ensuring high-quality garment inputs for virtual try-on. Lifestyle images made it difficult for the model to isolate clothing accurately. We improved results by using clean, white-background product images, which made the garment extraction and final renders much more consistent.

Wearwise shows how AI can make online fashion discovery more personal, interactive, and confident.