Abstract Expression recognition is a key issue in natural human-computer interaction. The machine leaning method is most commonly used to recognize expression from face images or image sequences, and good results are acquired. However, expression recognition is considered as a special process of human vision and it may provide a new way to realize emotion recognition in computer by simulating human emotional information process. The expression recognition using extended neural networks is presented. It simulates the structure of emotion-attention circuit in human brain by adding feedback control for attention in hidden layer. The effectiveness of the proposed approach is evaluated on Cohn Kanade expression database, and good results are obtained.
[1] Goleman D. Emotion Intelligence. New York, USA: Bantam Press, 1995 [2] Picard R W. Affective Computing. Cambridge, USA: MIT Press, 1997 [3] Fasel B, Luettin J. Automatic Facial Expression Analysis: A Survey. Pattern Recognition, 2003, 36(1): 259-275 [4] Moren J. Emotion and Learning [EB/OL]. [2002-10-26]. http://www.lucs.lu.se/People/Jan.Moren/Thesis.html [5] Taylor J G, Fragopanagos N, Cowie R, et al. An Emotional Recognition Architecture Based on Human Brain Structure // Proc of the 13th International Conference on Artificial Neural Networks and Neural Information Processing. Istanbul, Turkey, 2003: 1133-1140 [6] Zhou W, Coggins R. Computational Models of the Amygdala and the Orbitofrontal Cortex: A Hierarchical Reinforcement Learning System for Robotic Control // Proc of the 15th Australian Joint Conference on Artificial Intelligence. Canberra, Australia, 2002: 419-430 [7] Taylor J G, Fragopanagos N. ANNA: An Artificial Neural Network for Attention to Emotional Recognition // Proc of the 7th International Conference on Knowledge-Based Intelligent Information and Engineering Systems. Oxford, UK, 2003: 607-614 [8] Wang Shangfei, Xue Jia, Wang Xufa. Case-Based Facial Action Recognition Using IGA // Proc of the 1st International Conference on Affective Computing and Intelligent Interaction. Beijing, China, 2005: 80-87 [9] Ekman P, Friesen W V. Constants across Cultures in the Face and Emotion. Journal of Personality Social Psychology, 1971, 17(2): 124-129 [10] Xue Jia. Research on Several Key Technical Problems in Facial Expression Recognition. Master Dissertation. Hefei, China: University of Science and Technology of China. Department of Computer Science, 2005 (in Chinese) (薛 佳.面部表情识别若干关键问题研究.硕士学位论文.合肥:中国科学技术大学.计算机科学与技术系, 2005)
2008, Vol. 21 
