SG Wardrobe

About the Project

I've always struggled with deciding what to wear, changing in and out of outfits because nothing feels quite right, and my friends said the same. Fashion isn't one-size-fits-all either: what looks great on a model can look completely different on us, so buying clothes online often ends in disappointment and returns.

So I wondered: what if we could make the most of the clothes we already own, and check how new clothes will fit our exact bodies before we click “buy”?

What It Does

SG Wardrobe is a hyper-local AI personal stylist:

• smart closet and weather-aware styling — add your clothes and get complete outfits for the occasion, planned around Singapore's live weather, humidity, rain forecast and haze from the official Meteorological Service Singapore and NEA feeds, down to the freezing indoor aircon;
• body type and colour analysis — enter your measurements and skin tone to find your body shape and the palettes that flatter you;
• tailored recommendations — cuts and silhouettes that suit your proportions, plus a closet review of what stays, what to restyle and what goes; and
• Smart Shopping — paste a link to a clothing item, and the AI reads the product page and size chart, recommends your size, checks the fit at the bust, waist, hips and length, and scores how well it matches your wardrobe.

When a shop has no size chart, Smart Shopping says so and lowers its confidence instead of guessing. In the public demo, your closet and measurements stay in your browser.

Product Decisions

I first wanted a virtual try-on so you could see yourself in the suggested outfits, but none of the tools I tried could render it accurately without expensive enterprise APIs. I pivoted to deep measurement and silhouette analysis instead, which became the body type, colour analysis and Smart Shopping features.