Re-learning of Child Model for Misclassified data by using KL Divergence in AffectNet: A Database for Facial Expression

TL;DR

This paper proposes a re-learning approach using KL divergence to generate child models for misclassified facial expression images in AffectNet, improving classification accuracy for ambiguous cases.

Contribution

It introduces a novel method of creating child models for misclassified data based on KL divergence, enhancing emotion classification in deep learning models.

Findings

Child models improved classification of Disgust and Anger.

KL divergence effectively guides child model generation.

Re-learning reduces misclassification in ambiguous cases.

Abstract

AffectNet contains more than 1,000,000 facial images which manually annotated for the presence of eight discrete facial expressions and the intensity of valence and arousal. Adaptive structural learning method of DBN (Adaptive DBN) is positioned as a top Deep learning model of classification capability for some large image benchmark databases. The Convolutional Neural Network and Adaptive DBN were trained for AffectNet and classification capability was compared. Adaptive DBN showed higher classification ratio. However, the model was not able to classify some test cases correctly because human emotions contain many ambiguous features or patterns leading wrong answer which includes the possibility of being a factor of adversarial examples, due to two or more annotators answer different subjective judgment for an image. In order to distinguish such cases, this paper investigated a re-learning model of Adaptive DBN with two or more child models, where the original trained model can be seen as a parent model and then new child models are generated for some misclassified cases. In addition, an appropriate child model was generated according to difference between two models by using KL divergence. The generated child models showed better performance to classify two emotion categories: `Disgust' and `Anger'.