Research into automatic metadata collection of TV viewing trends of the elderly

In the context of the Call for the Creative Industries, TNO started the project MIME (Memory Banks Interactive Metadata Extraction). This project investigates the possiblility of registering the emotions, reactions and conversations of elderly viewers while watching television through cameras and sensors, and also automatically interpreting and linking these reactions to specific content to provide recommendations for the viewer.

This study aims to create an overview of possible methods that can be used for metadata extraction for and by elderly viewers in a fun and enjoyable way without impeding their social interaction. We deliberately chose to work with elderly viewers in this project and to study their TV viewing behaviours. Indeed, not only do they add valuable metadata to historical material, but they can also benefit from a system that helps them to recall old memories. Besides the metadata of television material, the effect of TV images on elderly is also studied in this project. Can TV footage motivate elderly to start being more active? Do people want to move more when viewing footage of their youth, or when viewing peers on TV? In addition to TNO, Ziggo and Noldus Information Technology BV made a financial contribution to this project.


Ir. Paul van den Haak

  • AI
  • big data
  • Computer Vision
  • National Safety
  • Sports & Media