作者: Günther Palm , Steffen Walter , Friedhelm Schwenker , Markus Kächele , Philipp Werner
DOI: 10.1007/978-3-319-20248-8_19
关键词: Modalities 、 Artificial intelligence 、 Fusion 、 Multiple classification 、 Facial expression 、 Computer science 、 Speech recognition 、 Feature selection 、 Experimental validation
摘要: In this work, multi-modal fusion of video and biopotential signals is used to recognize pain in a person-independent scenario. For purpose, participants were subjected painful heat stimuli under controlled conditions. Subsequently, multitude features have been extracted from the available modalities. Experimental validation suggests that cues allow successful recognition are highly similar across different people complementary analysed modalities an extent methods able achieve improvement over single Different approaches (early, late, trainable) compared on large set state-of-the art for biopotentials channels multiple classification experiments.