Emotion Categorization from Video-Frame Images Using a Novel Sequential Voting Technique

Abstract

Emotion categorization can be the process of identifying different emotions in humans based on their facial expressions. It requires time and sometimes it is hard for human classifiers to agree with each other about an emotion category of a facial expression. However, machine learning classifiers have done well in classifying different emotions and have widely been used in recent years to facilitate the task of emotion categorization. Much research on emotion video databases uses a few frames from when emotion is expressed at peak to classify emotion, which might not give a good classification accuracy when predicting frames where the emotion is less intense. In this paper, using the CK+ emotion dataset as an example, we use more frames to analyze emotion from mid and peak frame images and compared our results to a method using fewer peak frames. Furthermore, we propose an approach based on sequential voting and apply it to more frames of the CK+ database. Our approach resulted in up to 85.9% accuracy for the mid frames and overall accuracy of 96.5% for the CK+ database compared with the accuracy of 73.4% and 93.8% from existing techniques.

A comparison of humans and machine learning classifiers categorizing emotion from faces with different coverings

“…The learning rate was reduced by 10% after 80, 100, 120, 160 epochs, and by 5% after 180 epochs to avoid overfitting 1 the data. Note that the computer systems were trained with images from the last-half frames of the CK+ dataset based on the technique developed by Shehu et al [41].…”

Facial Expression Recognition Based on Multi-Channel Attention Residual燦etwork

“…Accuracy rate NSVT [45] 96.5 DRL [46] 89.8 CUDL [47] 96.6 CNN [48] 92.81 VGG16 [49] 94.8 HCIA [50] 9 6 DTAGN [51] 97.2 ST-RNN [52] 97.2 ResNet18 86.3 The proposed model 98.8…”

An anti-attack method for emotion categorization from images

“…In this research, a total of 3,368 images, which consists of the last half of the frames of each sequence of the six basic expressions are used as the peak expressions and the first two frames of each sequence are used as the neutral expression based on the technique developed by Shehu et al [64,65]. Fig.…”