作者: Markus Kächele , Patrick Thiam , Mohammadreza Amirian , Philipp Werner , Steffen Walter
DOI: 10.1007/978-3-319-23983-5_26
关键词:
摘要: In this work, a method is presented for the continuous estimation of pain intensity based on fusion bio-physiological and video features. The focus paper to analyse which modalities feature sets are suited best task recognizing levels in person-independent setting. A large set features extracted from available channels (ECG, EMG skin conductivity) stream. Experimental validation demonstrates contribute most robust prediction effects when combining them improve given unseen persons.