Academy of Finland
Funding decision
 
Name Zhao, Guoying
Organisation OY
Project title
Computer Vision for Continuous Emotional State Analysis
Decision No. 277395
Decision date 12.06.2014
Funding period 01.09.2014 - 31.08.2018
Funding 601 025
WebFOCUS Report
Project description
This project was to invent novel methodologies for analyzing continuous emotional states of a person from long facial video sequences. Image and video descriptors, continuous emotion and intensity, micro-expression analysis have been investigated to reach the goals. For descriptors, a general second-order pooling framework was formulated, and transferring prior knowledge from image domain to video domain was investigated. All have been used effectively for emotion analysis. For analyzing valence and arousal of emotions, new approaches for multi-modal video-induced emotion recognition based on facial expression and electroencephalogram (EEG) technologies, inferring the intensity of happiness and pains were proposed and developed. An improved spontaneous micro-expression dataset: CASME II was collected, and novel approaches for recognizing and detecting micro-expressions have been explored. With the carried research, affective computing has been moving further to dynamic and subtle.