Non-contact Digital Signage Allowing Users to Change Contents with Their Face Orientation

In today's society, digital signage is widely used in various places such as commercial facilities and train stations. Most digital signages use a liquid crystal display or LED panel. One of the advantages of digital signage is the ability to dynamically change the content of the advertisement. However, the method of change is not due to the viewer's intention, but rather to elapsed time to show, and it is merely a sequential display of static advertisements. In this study, we propose and implement a non-contact interactive digital signage system using a web camera that allows users to change the content according to their face orientation. In the implementation, we use OpenCV for showing the content and image processing, and the machine learning library, named dlib, is used for detecting user's face and identifying its orientation. The proposed system changes the layout and enlarges the content in which the user is interested. The user does not need to wear any device and to do any special action. Furthermore, the proposed system offers an intuitive interaction manner to the user. In this paper, some experimental designs are also described.