MULTIPLE INSTANCE LEARNING FOR HUMAN EMOTION ANALYSIS USING GABOR FEATURES

P Nagarajan

doi:10.29284/ijasis.4.2.2018.31-37

Abstract

Facial expression analysis (FEA) or Human Emotion Analysis (HEA) is an essential tool for human computer interaction. The nonverbal messages of humans are expressed by facial expression. In this study, an HEA system to classify seven classes of human emotions like happy, sad, angry, disgust, fear, surprise and neutral is presented. It uses Gabor filter for feature extraction and Multiple Instance Learning (MIL) for classification. Gabor filter analyzes the facial images in a localized region to extract specific frequency content in specific directions. Then, MIL classifier is used for the classification of emotions into any one of the seven emotions. The evaluation of HEA system is carried on JApanese Female Facial Expression (JAFFE) database. The overall recognition rate of the HEA system using Gabor and MIL technique is 95%.

Highlights

The human beings express their emotions, intensions and other nonverbal messages by facial expression
All the images are considered by Human Emotion Analysis (HEA) system for the performance evaluation
The overall accuracy of HEA system is 92.5% which shows that the Gabor features extracts dominant features to classify the emotions successfully with the help of Multiple Instance Learning (MIL) classifier

Summary

INTRODUCTION

The human beings express their emotions, intensions and other nonverbal messages by facial expression. SVM model is used for classification of facial expression images. SVM classifier is used for emotion recognition using facial features. Deep extreme extraction network based facial expression classification is described in [6]. The input facial expression images are classified by the deep convolutional neural network based Xception model. Quantum Neural Network (QNN) based feature difference matrix for facial expression recognition is described in [8]. Sparse representation based facial expression recognition is described in [9]. Binarized statistical image features based facial expression recognition is discussed in [11]. SVM ensemble classification for facial expression recognition is described in [12]. The recognition of facial expression is discussed in [14] using SVM and geometric deformation features.

METHODS AND MATERIALS

Gabor Filter

MIL Classification

RESULTS AND DISCUSSION

80 Anger Surprise Disgust Sad

CONCLUSION