Understanding viewers' involuntary behaviors for adaptive digital signage

Abstract

Digital signage has been getting more popular due to the recent development of underlying hardware technology and improvement in installing environments. In order to deliver a message from a sender effectively, it is necessary to check up viewers' evaluations towards the content actually displayed by the digital signage and utilize the evaluations to make it more adapted to the local viewers' tastes. However, previous methods for the survey such as interviewing require a lot of human resources, and thus there have been some research attempts to gather viewers' evaluations towards digital signage automatically. For example, some digital signage systems utilize a fixed video camera for the evaluation. They recognize viewers' faces, and understand their demographics segmentation, such as gender and age, though these values are not related to their actual feelings towards digital signage.