Video Analytics for Retail

Overview:

    • 2017-18, partnered w/ R&D at Xylinx, Intel, Google, and XNOR for next generation retail smart camera design
    • Demo’d initial ‘back end’ AI/ML models interactively at NRF annual conference in NYC in 2019.
    • Then deployed intel solution as pilot in San Jose March 2019
      • 14 streaming cameras w/ ensemble (object detection, tracking)
      • goal was to measure throughput of 4 sales channels (online order w/ pickup, drive-thru)

Pilot Summary:

  • object classification and tracking in a commercial setting: Vehicles and people.
  • Then re-id across cameras was added, and then some demographics and other KPI for interest to retail for final AI/ML pipeline.

 

Also see an Intel Blog on some of the outcomes of the R&D efforts

 

Tools : OpenVino , C++, Python, FPGA

following images roughly in order from planning to prototype to pilot.

planning for smart camera R&D

working AI/ML video pipeline prototype

KPI Breakdown of Next Generation Smart Camera for Retail

Exhibition Booth Camera Placement

Exhibition Booth Segmented Traffic from Video AI/ML

object detection, tracking, and re-id pilot