NCKU professor’s research wins thesis award competition
Tainan, Taiwan, January 5, 2014
A Tainan-based research team led by Dr. Chung-Hsien Wu, a professor of computer science and information engineering from National Cheng Kung University (NCKU) has developed a hybrid fusion method for achieving high-performance emotion recognition result based on audio-visual bimodal fusion.
The study recently has won the Golden Award in the Fifth Merry Electroacoustic Thesis Award Competition, sponsored by Merry Electronics, announced the university.
Dr. Wu working with his PhD student Mr. Jen-Jun Lin for many years has made a breakthrough in research and their study titled “A Study on Data Fusion Strategy for Audio-Visual Emotion Recognition,” received recognition for outstanding work.
Dr. Wu said, “Recent years have seen increased attention being given to research topic in automatic audio-visual emotion recognition for natural human-computer interaction.”
He explained, the study propose a novel model-level fusion method with respect to the temporal course modeling scheme to effectively solve the problem of complex temporal structures of emotional expressions for audio-visual emotion recognition in a conversational environment.
This technology could be applied to affective robot, intelligent classroom, intelligent home, games, emotion-based music retrieval, and negative emotion detection for healthcare, according to Dr. Wu.
Enditem/
A Tainan-based research team led by Dr. Chung-Hsien Wu, a professor of computer science and information engineering from National Cheng Kung University (NCKU) has developed a hybrid fusion method for achieving high-performance emotion recognition result based on audio-visual bimodal fusion.
The study recently has won the Golden Award in the Fifth Merry Electroacoustic Thesis Award Competition, sponsored by Merry Electronics, announced the university.
Dr. Wu working with his PhD student Mr. Jen-Jun Lin for many years has made a breakthrough in research and their study titled “A Study on Data Fusion Strategy for Audio-Visual Emotion Recognition,” received recognition for outstanding work.
Dr. Wu said, “Recent years have seen increased attention being given to research topic in automatic audio-visual emotion recognition for natural human-computer interaction.”
He explained, the study propose a novel model-level fusion method with respect to the temporal course modeling scheme to effectively solve the problem of complex temporal structures of emotional expressions for audio-visual emotion recognition in a conversational environment.
This technology could be applied to affective robot, intelligent classroom, intelligent home, games, emotion-based music retrieval, and negative emotion detection for healthcare, according to Dr. Wu.
Enditem/
Click Num:
Share
