Recommendation Engines and Visual Search

Understanding consumer behavior has never been more critical for retailers. To drive growth, they’re using personalized recommendations and augmented reality (AR) environments to create tailored experiences and improve revenue. 

We're using AI to simplify our customer experience. In general, retailers are using AI to optimize prices by balancing demand and supply, analyzing the performance of discount programs and sales, and setting prices that work for the business and customers, all while responding to real-time market changes.

— Victoria Uti, Director, Principal Research Engineer, Kroger

Pricing optimization helps to predict the impact of the changes in price, the likely demand at those prices, and the best recommendations to choose from. AI can play a vital part in the process where, traditionally, maybe a merchandiser would have to review every single pricing recommendation that's being made across thousands of stores and potentially millions of products.

— Rob Armstrong, Director of Data Science, Tesco

Recommendation Systems

On some of the largest commercial platforms, recommendations account for as much as 30 percent of revenue, which translates into billions of dollars in sales. That’s why retailers are using AI recommender systems to drive every action shoppers can take, from visiting a webpage to using social media for shopping. They also improve conversion by offering relevant products from the exponential number of available options. 

NVIDIA Merlin, an end-to-end recommender system framework, provides rapid extract, transform, and load (ETL) functions, feature engineering, and high training throughput to enable fast experimentation and production retraining of recommender models. Merlin also enables low-latency, high-throughput, production inference.

Personalized Recommendations

To engage consumers, retailers need to deliver on an expectation of one-to-one personalization. 

Olay Skin Advisor, a GPU-accelerated AI tool that works on any mobile device, assesses a user-provided selfie and recommends an Olay regimen to improve trouble areas. After four weeks, 94 percent of users continued to apply the recommended products. 

Stitch Fix, a fashion ecommerce company, provides a seamless balance between AI-powered decision making and human judgment. By using algorithms to understand customer preferences, Stitch Fix’s fashion service combines the art of personal styling with data analytics—all powered by GPU-accelerated deep learning.

Autotagging

Retailers are using next-generation computer vision to recognize image attributes and automatically generate meta-tagging and cataloging for search and recommendation results. This comprehensive information about products and services results in successful recommendation personalization. 

Since fashion changes quickly, NVIDIA partner Omnious offers an AI-tagging API that helps customers stay ahead of the fashion curve. Ominous Tagger, the automated tagging solution with over 95 percent accuracy, is 100X faster than manual tagging and increases search efficiency by 4X. Omnious also offers a trend report that analyzes social media images from fashion influencers.

Virtual Fitting

The 2021 cost of returned merchandise in the US was $761 billion. Online returns accounted for $218 billion of that total. To reduce the number of returns and provide a more enhanced shopping experience, retailers can now suggest items to customers that are virtually guaranteed to fit.

Cappasity’s Virtual Try-On solution enables customers to virtually try on garments to see how they look before they buy. Powered by NVIDIA GPUs, with CUDA® to boost the speed of calculations, Cappasity’s algorithms process data in the cloud to detect body measurements, while neural networks perform human contouring segmentation.

Sign up for the latest retail news from NVIDIA.