Drape

• 

Inspiration

Choosing an outfit is more than finding something that looks good on a model. The right outfit depends on skin tone, colour harmony, occasion, personal style, and budget. We wanted to build an experience that answers a simple question:

“What would actually look good on me?”

This inspired Drape, an AI-powered personal styling experience that creates a complete look personalized to the individual—not just an outfit recommendation.

What it does

Drape creates a personalized styling journey:

Upload → Analyze → Recommend → Try On → Complete the Look

• Users select their occasion, preferred style, outfit type, and budget.
• Users upload their photo for skin-tone analysis.
• Drape recommends the 3 best outfits based on skin tone, preferences, occasion, and budget.
• Users select their favourite outfit and virtually try it on using YouCam.
• After the virtual try-on, Drape recommends a hairstyle and lipstick shade that complement both the user's skin tone and selected outfit.
• The result is a complete personalized look, rather than simply a clothing recommendation.

How we built it

We built Drape around the YouCam API, combining virtual try-on with our own personalization and recommendation experience.

The recommendation flow considers:

• Skin tone and undertone
• Outfit colour compatibility
• Occasion
• Personal style
• Outfit preference
• Budget

The selected outfit is then connected to its corresponding garment reference and passed through the YouCam virtual try-on experience.

We also created a structured product catalog containing outfit attributes such as colour, style, occasion, price, and suitability, allowing Drape to rank and recommend outfits instead of simply displaying random products.

After the try-on, the selected outfit becomes the basis for the final beauty recommendations—creating a cohesive outfit + hairstyle + lipstick combination.

Challenges we ran into

One of our biggest challenges was making the recommendations feel personalized rather than random. Matching an outfit requires considering several factors simultaneously, especially skin tone, colour compatibility, occasion, style, and budget.

Another challenge was connecting the recommended catalog item with the correct YouCam garment reference, so that the outfit the user selects is the same outfit they see during virtual try-on.

We also had to balance the scope of the project with limited hackathon time. Instead of building a complete fashion marketplace, we focused on delivering one polished end-to-end experience.

Accomplishments that we're proud of

We are proud that Drape goes beyond “find an outfit” and creates a complete styling experience.

The user can go from:

Their photo → Personalized outfit recommendations → Virtual try-on → Hairstyle + lipstick → Complete look

We are especially proud of combining personalized colour-based styling with virtual try-on, because it allows users to not only receive a recommendation but actually visualize themselves in the selected outfit.

What we learned

We learned that meaningful personalization requires more than matching a user's preferences to a product.

A good styling recommendation needs to understand the relationship between the person, outfit, and overall look.

We also learned how important it is to separate catalog/product images from virtual try-on garment assets, since they serve completely different purposes.

Most importantly, we learned how to turn a broad idea into a focused product experience by prioritizing the features that create the most value for the user.

What's next for Drape

Drape can evolve into a complete AI personal stylist.

Future possibilities include:

• 💎 Earrings, bracelets, shoes, and bags
• 👗 More advanced body-proportion and fit recommendations
• 🛍️ Shopping recommendations from platforms such as Amazon and Myntra
• 🔗 Direct product and purchase links
• 👚 Personalized wardrobe recommendations

Built With

• react
• typescript
• youcam