VICAR - Visual InStore Customer Analytics and Recommendations

  • Duration:

The project "Visual Instore Customer Analytics and Recommendations" (VICAR) pursues the goal of providing stationary retailers with information about customer journeys and offering value-added services for retailers and customers based on this information. Within the project, a software tool is being developed that uses innovative machine learning technologies to analyze video data and thus predict the individual shopping paths of customers. Here, the video data is extracted from existing video surveillance for theft protection. Thus, by analyzing the real-time data, it is possible, for example, to identify customers who are in need of assistance or to display individualized advertisements to them. In addition, conspicuous movement patterns can be used to detect shoplifters and, if necessary, inform security personnel.


  • Schirra IT GmbH - IS Predict GmbH - August-Wilhelm Scheer Institut für digitale Produkte und Prozesse gGmbH - Deutsches Forschungszentrum für Künstliche Intelligenz GmbH

Assoziierte Partner: - Horn & Company Data Analytics GmbH

Share project:

Contact Person
Frederic Kerber, M.Sc.

Publications about the project

Julian Groß, Marcel Köster, Antonio Krüger

In: Computer Graphics & Visual Computing (CGVC) 2019. Computer Graphics and Visual Computing (CGVC-2019) September 12 Bangor Wales United Kingdom The Eurographics Association 2019.

To the publication

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz