Subtle Expression Recognition Using Optical Strain Weighted Features

• Published in: Computer Vision - ACCV 2014 Workshops pp. 644 - 657
• Main Authors: Liong, Sze-Teng, See, John, Phan, Raphael C.-W., Le Ngo, Anh Cat, Oh, Yee-Hui, Wong, KokSheik
• Format: Book Chapter
• Language: English
• Published: Cham Springer International Publishing 2015
• Series: Lecture Notes in Computer Science
• Subjects: Emotion Recognition; Facial Expression; Local Binary Pattern; Optical Flow; Support Vector Machine
• ISBN: 9783319166308; 3319166301
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-319-16631-5_47

Optical strain characterizes the relative amount of displacement by a moving object within a time interval. Its ability to compute any small muscular movements on faces can be advantageous to subtle expression research. This paper proposes a novel optical strain weighted feature extraction scheme for subtle facial micro-expression recognition. Motion information is derived from optical strain magnitudes, which is then pooled spatio-temporally to obtain block-wise weights for the spatial image plane. By simple product with the weights, the resulting feature histograms are intuitively scaled to accommodate the importance of block regions. Experiments conducted on two recent spontaneous micro-expression databases–CASMEII and SMIC, demonstrated promising improvement over the baseline results.