Unique Impression Recognition in a Non-Obtrusive, Non-Invasive and Privacy-Preserving Manner

Funding Details
Natural Sciences and Engineering Research Council of Canada
  • Grant type: Engage Grants Program
  • Year: 2014/15
  • Total Funding: $25,000
Principle Investigator(s)

No researchers found.


Project Summary

Advolve, is a leader in projecting information content on a two-way glass mirror. This has allowed thecompany to turn its innovative display platform for projecting advertisement content in ubiquitous settings. AnIR proximity sensor detects a viewer's presence, in locations where mirrors are commonly used, such as awashroom vanity, and the system displays advertisement in correspondence to such forms of user engagement.The platform is connected to the internet, thus providing real-time feedback to a back-end server for analyticsto gather information about which ads were played, where and when, along with engagement information aboutthe viewer. Advertisers place tremendous value in understanding how many unique impressions were affordedby their ads. Therefore, to cater to this critical requirement, Advolve needs to advance its technology so that itcan identify all unique users who engage with the system to boost its analytics capabilities. This researchentails identifying a suitable algorithm to provide unique characteristics of individual viewers, withoutcompromising their true identity. Since ads can be viewed in virtually any location (such as in washrooms),user identity cannot be recorded using any form of imaging system. Instead proxy information is necessary toencode unique impressions. The team consists of researchers at the University of Manitoba with expertise inhuman-computer interactions well as experts from Advolve in video and image processing. The outcome willinvolve developing a proof-of-concept system to facilitate user identification in an non-obtrusive, non-invasiveand privacy-preserving manner. As a result of this development, Advolve will have access to new technologyallowing them to uniquely identify users of their displays.

Related Projects

Data is growing relentlessly. As such, individuals and organizations are relying more so than ever on extracting meaning from large data repositories. Managing gigantic data flows, probing massive repositories and speedily extracting value is becomin... More ...
Visual analytics, "the science of analytical reasoning facilitated by the interactive visual interface", seeks to develop an evidenced-based approach to the design and evaluation of cognitive systems comprising one or more human decision-makers suppo... More ...