Inspiration

An enormous problem within retail stores is the friction in the process of customer service. For example, every single time a customer enters a T-Mobile store, the representative must tediously record his/her profile. This causes disengagement from the customer and an increase in customer wait time. They need a more seamless approach to accommodate customers and right-fit them to T-Mobile's most appropriate products and services. We aim to solve this problem with AI/ML technologies.

What it does

  • Identifies customers that come into retail stores with face recognition and provides an automatic sign in to the representatives' devices.
  • Sends followup email.
  • Manages promotional deals based on prior customer activity.

How we built it

We used Azure Face API and DLIB face recognition to accurately identify incoming customers into the store. Website built using Angular to map the faces to existing T-Mobile IDs.

Challenges we ran into

We encountered some issues using Azure's Face API when identifying faces, specifically with the endpoints.

Accomplishments that we're proud of

A fully working DEMO.

What we learned

How to integrate complex servers together. Using ML APIs proved to be very challenging, and in addition, some may constrain you on resource usage.

What's next for FaceIO

Build infrastructure to support a scalable solution. Store face recognition in-house for an unlimited number of face recognition calls. Switching to a hosting provider.

Share this project:

Updates