Facial Expression Recognition using Patch based Gabor Features

Abstract

Facial Expression is one of the most natural, and powerful means for human beings to Communicate their emotions and intentions. The recognition of facial expressions is very important for interactive Human Computer Interfaces. One crucial step for facial expression recognition (FER) is the accurate extraction of emotional features. Numerous feature extraction techniques have been developed for recognition of expressions from static images as well as videos. This paper put forward an approach using distance features that are obtained by extracting patch based 3D Gabor features and conducting patch matching operations. The experimental results shows high correct recognition rate (CRR), fast processing time and significant performance improvements because of the consideration of facial components and muscle movements. Comparison with the state-of-the art indicates that the proposed approach achieves high CRR for JAFFE database and is one among the top performers on the Cohn-Kanade (CK) database.

A Survey of AI-Based Facial Emotion Recognition: Features, ML & DL Techniques, Age-Wise Datasets and Future Directions

“…Gabor features have also been combined with a patch-based extractor to overcome the lack of accuracy on the linear representation of the small sample size [115]. Another study [116] used this combination where the 3D Gabor features and patch method were used. In some cases, the extracted patches of images are converted into an image matrix in a PCA framework using the patch method.…”