Bringing Computer Vision to the 'edge' | Blog | Trax

Bringing Computer Vision to the ‘edge’

Bringing Computer Vision to the ‘edge’

Trax and Google have partnered to bring Trax’s breakthrough image recognition capability to every retail store allowing retailers to take advantage of real-time shelf insights at unprecedented speed and scale with increased security and privacy.

Trax will use Google’s revolutionary Cloud IOT platform and Edge TPU to enhance Trax Retail Watch, a store monitoring and intelligence solution that uses Internet of Things (IoT) devices to help retailers address shelf gaps.

  • Edge TPU,  a new purpose-built hardware accelerator ASIC chip, installed in Trax Gateway Computer, extends Trax‘s powerful image recognition capability to stores locally using pre-trained models. It significantly reduces latency and increases the speed to analyze shelf images, captured by shelf-mounted cameras
  • Edge IOT Core, a component of Google Cloud IOT Edge, securely connects Trax Gateway Computer to Trax Cloud, enables software and firmware updates, and manages the exchange of data. This allows connected devices to act on the data they receive in real-time and deliver the most accurate results

What's in it for retailers?

  • Faster real-time results at scale: By running on-premise image recognition and analyzing shelf conditions in real-time, we provide significantly faster insight into on-shelf availability and planogram compliance levels of connected shelves, delivered at highest availability and at scale.
  • Enhanced operational reliability: Our solution allows retailers to locally process shelf images and receive actionable insights from connected devices without worrying about connectivity issues in the store
  • Increased security for devices and data: By analyzing shelf images locally on edge devices instead of sending raw data to the cloud, the solution helps retailers eliminates the need to send large and potentially sensitive data to the cloud

By implementing Trax Retail Watch, retailers will be able to reduce operational costs, increase revenues, and secure additional analytics and insights about the shelf to support both immediate and future in-store strategies.

Related posts

by Kata Kovari

Traxer in the Spotlight: Iris Guzmán has an insatiable appetite for meeting new people, trying new food and experiencing new cultures!

We catch up with Operations Service Manager Iris Guzmán who has an insatiable appetite for meeting new people, trying new food and experiencing new cultures!

Read more
by Connie Cheng

How to optimize your category assortment using shelf data

There are two dimensions to product assortment: product breadth and product depth. The number of different lines kept in stock determines product breadth, and product depth measures the number of units of each line.

Read more
Back to top