The Effect of the Number of Key-Frames on the Facial Emotion Recognition Accuracy

Abstract

Key-frame selection plays an important role in facial expressionrecognition systems. It helps in selecting the most representative framesthat capture the different poses of the face. The effect of the number ofselected keyframes has been studied in this paper to find its impact on thefinal accuracy of the emotion recognition system. Dynamic and staticinformation is employed to select the most effective key-frames of thefacial video with a short response time. Firstly, the absolute differencebetween the successive frames is used to reduce the number of frames andselect the candidate ones which then contribute to the clustering process.The static-based information of the reduced sets of frames is then given tothe fuzzy C-Means algorithm to select the best C-frames. The selectedkeyframes are then fed to a graph mining-based facial emotion recognitionsystem to select the most effective sub-graphs in the given set of keyframes.Different experiments have been conducted using Surrey Audio-VisualExpressed Emotion (SAVEE) database and the results show that theproposed method can effectively capture the keyframes that give the bestaccuracy with a mean response time equals to 2.89s