作者: Robert Niese , Ayoub Al-Hamadi , Bernd Michaelis
DOI: 10.1007/978-3-642-31298-4_8
关键词: Pose 、 Face hallucination 、 Pattern recognition (psychology) 、 Facial expression 、 Three-dimensional face recognition 、 Computer science 、 Realization (linguistics) 、 Field of view 、 Computer vision 、 Adaptation (computer science) 、 Artificial intelligence
摘要: In his paper we describe a multi-camera based concept for facial expression recognition, which is of potential interest modern man-machine-interfaces. Naturally, the recognition user events limited by field view applied cameras. However, in some application domains, such as patient state analysis it mandatory to always get feedback. This can only be accomplished increasing observable view. Our proposed addresses this issue through use multiple For realization extended an existing technique. Examples are given three camera setup, substantially enhances degree freedom interaction and/or events. article components, adaptation generic face model, pose estimation and relevant feature transformations carry out machine recognition.