Browse Prior Art Database

Cognitive and Contextual Custom Push Notifications Using Video Analytics

IP.com Disclosure Number: IPCOM000246289D
Publication Date: 2016-May-24
Document File: 2 page(s) / 67K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method and system to create custom notifications for customers using contextual and cognitive resources based the analyzed video feed metadata.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 2

Cognitive and Contextual Custom Push Notifications Using Video Analytics

The novel contribution is a method and system to create custom notifications for customers using contextual and cognitive resources based the analyzed video feed metadata. This novel method and system enablestores to push notifications to the optimal subset of people. In addition, the system can identify the physical features of a person associate those features with a device, and save the information.

A smart location awareness system with associated video analytics is set up in a venue (e.g., retail establishment). This requires hardware and software to be set up with venue information.

Figure 1: Smart location awareness system set up in a retail venue

Figure 1 illustrates a venue with Wi-Fi sensors set up for triangulating a device's position in the store. A video camera is watching the store entrance area. To provide location and direction context, the venue places a camera in the store (according to a floor plan), which feeds the smart location awareness system. A Wi-Fi sensor on the floor plan to enables triangulation. (The technology that enables a correlation between a person tracked in a video feed and a device is enabling art and is not novel.)

To implement the system to generate cognitive and contextual custom push notifications using video analytics:

1. A person walks into a venue 2. A video camera captures a person walking in the venue 3. The system passes the video feed through analyzing software to extract metadata (e.g., person is wearing a hat, black jacket, jeans,...