AI Based Personal Shopping Assistant

Oyper will help consumers make better fashion decisions. Oyper technology builds fashionable outfits and provides personalized fashion recommendations.

HIGHLIGHTS

Features you love

We are currently working on three products

Our project is coming soon!

Style DNA

Provides access to millions of unique interactive outfits developed specifically for you by intelligent algorithms.

Find Similar

Just upload an image with an outfit that you like, and Oyper will tell you where to buy a similar outfit at your selected price point.

Closet

Oyper shows you Untapped Potential of your Closet. Review multiple ideas of how to mix and match items in your close and new pieces.

CUTTING-EDGE

Oyper Technology

Artificial Intelligence

Oyper algorithm is powered by machine learning and computer vision. The algorithm learns to understand fashion trends in real time and develops targeted recommendations tailored to unique consumer styles.

Computer vision algorithms

The algorithm understands fashion trends and styles. Identifies textures, colors, materials and many other attributes and matches these attributes based on patterns and combinations observed in fashion blogs, websites, and other fashion publications

Machine learning

Oyper uses multiple technologies to understand the style and your preferences. We analyze the results of the items you like, so the more you use it, the better it gets.

Recommendation algorithm

Oyper learns about the consumer’s style and preferences based on how the customer behaves on the website, including styles reviewed, clicks, uploaded images, amount of time spent on each recommended look and modifications to recommended looks implemented by the customer.

Big data processing

There is enormous amount of data available and we believe by analyzing at least a small part of this data we are building an intelligent machine which can replace and automate fashion styling and shopping for cloth.

Real-time trend analytics

Oyper’s understanding of fashion comes from continuous analysis of outfits which people wear on social networks, fashion websites and personal fashion blogs.

INFO

Learn more about us

graphic

We are developing Oyper, an AI Based Personal Shopping Assistant

Oyper will help consumers make better fashion decisions. Oyper technology builds fashionable outfits and provides personalized fashion recommendations.

Oyper algorithm is powered by machine learning and computer vision. The algorithm learns to understand fashion trends in real time and develops targeted recommendations tailored to unique consumer styles.

Oyper’s understanding of fashion comes from continuous analysis of outfits which people wear on social networks, fashion websites and personal fashion blogs. Based on this knowledge, Oyper builds outfits by matching various styles, clothes, colors, and materials. Once an individual consumer starts using the website, Oyper learns about the consumer’s style and preferences based on how the customer behaves on the website, including styles reviewed, clicks, uploaded images, amount of time spent on each recommended look and modifications to recommended looks implemented by the customer.

We have discovered a unique approach to solving the problem of building fashion outfits using AI.

We have a working demo which demonstrates that our technology understands the following:

1. The algorithm understands fashion trends and styles and recommends outfits which follow a particular style or fashion trend

2. Identifies clothing pieces based on visual recognition: bottom, top, dress, shoes, accessories and matches items correctly

3. Identifies textures, colors, materials and many other attributes and matches these attributes based on patterns and combinations observed in fashion blogs, websites, and other fashion publications

4. Understands specific nuances of styles and cuts - can identify unique features of the product - symmetry / asymmetry, high shoulder / low shoulder, high turtle neck / low turtleneck, skinny pants / regular / wide etc

5. Very high precision level which is a result of continuous research, refinement of our algorithm and expansion of our dataset to teach the machine

graphic
graphic

Fully automated AI Based Personal Shopping Assistant

Key new feature of the technology we offer is the ability to build fashionable outfits without the involvement of a human stylist.

Our objective is to build a fully automated AI Based Personal Shopping Assistant that generates outfit ideas and recommendations. We have developed a fully automated proprietary approach to building outfits using AI and computer vision to provide personalized and targeted recommendations.

People currently use the following solutions:

1. Fashion aggregators - to be able to look through selections from many brands on one website but there is no styling component involved

2. Pinterest - people save ideas on Pinterest and look where to find similar items

3. Stitch Fix - provides outfit recommendations based on input from human stylists, in process of building AI technology

4. Fashion blogs and social media - customers review various ideas to get inspired for building their outfits

We discovered our proprietary approach

We think it is a time to exclude human involvement from online shopping assistance.

There is enormous amount of data available and we believe by analyzing at least a small part of this data we can build an intelligent machine which can replace and automate fashion styling and shopping for cloth.

We have developed an algorithm that can build outfits by analyzing a bunch of images to copy the style and then apply it to new items generated from scratch.

We have done a lot of research in the field while developing the algorithm. Tried different state-of-the-art approaches to solve the problem, such as a tool called generative adversarial network (GAN). The GAN is a new technique in AI research that forces two neural networks against one another, using the outcome to improve the overall system. It learns from raw data — so by looking at a batch of images, it can draw out a style. That could mean spotting a particular trend on Instagram, Facebook, or its own collection of fashion imagery. But as a result of this work we discovered our proprietary approach. Our models are accurate and generate the most valuable results. We want Oyper to be “the best place to buy fashion online”.

graphic
dual phone

Style DNA

dual phone

Provides access to millions of unique interactive outfits developed specifically for you by intelligent algorithms.

Find Similar

dual phone

Just upload an image with an outfit that you like, and Oyper will tell you where to buy a similar outfit at your selected price point.

dual phone
dual phone

Closet

dual phone

Oyper shows you Untapped Potential of your Closet. Review multiple ideas of how to mix and match items in your close and new pieces. Helps you reinvigorate your closet and makes shopping fast and fun.

Launch

We are working on the improvement of our product for public launch which is expected in August.

Contacts us
dual phone

If you are interested in private demo you can always contact us, otherwise we’ll be online shortly!

Minsk, Belarus