ABSTRACT Identifying the Addressee in Human-Human-Robot Interactions determined by Head Pose and Speech (2008) Michael Katzenmaier
Abstract In this function we look into the energy of acoustic and visual cues,
Office 2007 Key, and their blend,
Microsoft Office 2010 Professional Plus, to identify the addressee in a human-human-robot interaction. According to eighteen audiovisual recordings of two human beings along with a (simulated) robot we discriminate the interaction from the two humans from your interaction of 1 human with the robot. The paper compares the consequence of 3 approaches. The initial strategy utilizes purely acoustic cues to locate the addressees. Very low level,
Microsoft Office 2010 Key, feature based mostly cues as well as higher-level cues are examined. Inside the second approach we examination whether the human’s head pose is actually a ideal cue. Our results present that visually believed head pose is a far more reliable cue for your identification from the addressee inside the human-human-robot interaction. From the third strategy we merge the acoustic and visual cues which results in substantial enhancements.
Details der Publikation Download Quelle Mitarbeiter CiteSeerX Archiv CiteSeerX - Scientific Literature Digital Library and Lookup Engine (United states of america) Key phrases attentive interfaces,
Office 2007, focus of interest,
Windows 7 Product Key, head pose estimation Typ text Sprache Englisch