|
Abstract
|
|
|---|---|
| Interactive displays are permeating public spaces at a rapidly accelerating pace. Research on how the success of a public display installation or application can be determined is rather scare and commonly accepted key performance indicators are missing. To address this challenge, several interaction models have been developed. Of particular interest is the Audience Funnel, which describes different stages of a process that ultimately leads to that people start interacting with a display. The aim of the project is to develop a tool that allows different phases of the interaction process to be determined through Microsoft Kinect and camera. The features of interest of pedestrians are their speed of walking, orientation of shoulder, head pose and gaze direction, which are indicators of attention of passersby to the display. These features can feed the Audience Funnel model and determine different stages of interaction, and then evaluate and improve the display installation or application. During the stay, my work was focused on the extraction of features of pedestrian?s walking speed, orientation of the shoulder and head pose with Kinect. | |
|
International
|
Si |
|
Place
|
University of Munich |
|
Type
|
|
|
Start Date
|
31/03/2014 |
|
End Date
|
29/06/2014 |