Facial expression contains a rich variety of affective information. Facial Expression Recognition (FER) is an emerging field that has broad applications in the fields of human-computer interaction, psychological behaviour analysis and image understanding. However, FER presently is not fully realized due to the lack of an effective facial feature descriptor. This paper will examine the performance of a novel facial image descriptor referred to as Local Directional Covariance Matrices (LDCM). The descriptor will consist of fusing features, such as location, intensity, filter responses and incorporate local texture patterns. Tests were done on both posed and spontaneous facial expression datasets to evaluate the performance of the proposed model on real-world applications. Results demonstrate the effectiveness of the covariance feature descriptors compared to standard methods.
Reference:
Naidoo, A., Tapamo, J.R. and Khutlang, R. 2018. Affective Computing: Using Covariance Descriptors for Facial Expression Recognition. International Conference on Image and Signal Processing (ICISP 2018), 2-4 July 2018, Cherbourg, France
Naidoo, A., Tapamo, J., & Khutlang, R. (2018). Affective Computing: Using Covariance Descriptors for Facial Expression Recognition. Springer. http://hdl.handle.net/10204/10397
Naidoo, A, JR Tapamo, and Rethabile Khutlang. "Affective Computing: Using Covariance Descriptors for Facial Expression Recognition." (2018): http://hdl.handle.net/10204/10397
Naidoo A, Tapamo J, Khutlang R, Affective Computing: Using Covariance Descriptors for Facial Expression Recognition; Springer; 2018. http://hdl.handle.net/10204/10397 .
Copyright: 2018 Springer.Due to copyright restrictions, the attached PDF file contains the accepted version of the paper. For access to the published version, please consult the publisher's website.