Dozens of shades, zero testers: we built Sephora's virtual try-on.

Sephora is a beauty retailer offering a wide selection of makeup, skincare, hair and fragrance products from both classic and emerging brands.

For Sephora's marketing team, Wildstream developed a computer vision and augmented reality system that overlays cosmetics on a live camera image, so customers can try dozens of options instantly.

Sephora
  • AI Engineering

Dozens of shades

Customers instantly test dozens of cosmetic options on a live image of themselves, without applying a single product

Live camera overlay

Augmented reality places products onto the live camera image, so the result moves naturally with the customer

Marketing-led idea

Sponsored by Sephora's Director of Marketing, the try-on turns product discovery into an engaging brand experience

AI Engineering

AR Virtual Try-On

Customers want to test many cosmetic shades before buying, which is impractical in store or online. They often want to compare several shades side by side before they decide. Testing each one physically takes time in store and is impossible when shopping online.

Dozens of shades

Customers instantly test dozens of cosmetic options on a live image of themselves, without applying a single product.

Live camera overlay

Augmented reality places products onto the live camera image, so the result moves naturally with the customer.

Marketing-led idea

Sponsored by Sephora's Director of Marketing, the try-on turns product discovery into an engaging brand experience.

What we did

Computer vision and augmented reality overlay cosmetic products on a live camera image, letting customers try dozens of options instantly.

Wildstream developed a computer vision and augmented reality system that overlays cosmetic products onto live camera images. Computer vision follows the customer's face in the camera image while augmented reality renders the selected product on top, letting customers switch between shades in a moment.

How it works
  1. The customer opens the live camera view
  2. Computer vision locates the face in the image
  3. The customer chooses a product and shade
  4. Augmented reality overlays the product on the live image
  5. The customer switches between options instantly
Key capabilities

Face tracking

Computer vision locates and follows facial features in the live camera image, keeping each product in the right place.

Real-time AR rendering

Augmented reality draws the chosen cosmetic onto the camera image in real time, so the preview follows every movement.

Instant shade switching

Customers move from one shade to the next immediately, comparing several options side by side within a few seconds.

Technology
  • Computer vision
  • Augmented reality
Impact

Dozens of cosmetic shades, tried on instantly through a live camera image

More confident choices

Trying many shades before buying helps customers choose products they are happy with, with fewer disappointing purchases.

Engaging shopping

Playful virtual testing keeps customers exploring the range longer and makes discovering new products part of the fun.

No shared testers

Customers can try products without touching shared testers, which makes testing many shades more comfortable and hygienic.

Bring us a challenge like this

Tell us where your operations still depend on manual work, and we will tell you what is worth building first.