Recognizing Facial Expressions in the Wild using Multi-Architectural Representations based Ensemble Learning with Distillation

TL;DR

This paper introduces EmoXNet, an ensemble learning model, and EmoXNetLite, a distilled, efficient neural network for facial expression recognition, achieving high accuracy and good generalization on benchmark datasets.

Contribution

The paper presents a novel ensemble model and a distillation approach for real-time facial expression recognition, improving accuracy and efficiency.

Findings

EmoXNet achieved 85.07% accuracy on FER2013.

EmoXNetLite achieved 82.07% accuracy on FER2013.

Models generalize well on new data.

Abstract

Facial expressions are the most common universal forms of body language. In the past few years, automatic facial expression recognition (FER) has been an active field of research. However, it is still a challenging task due to different uncertainties and complications. Nevertheless, efficiency and performance are yet essential aspects for building robust systems. We proposed two models, EmoXNet which is an ensemble learning technique for learning convoluted facial representations, and EmoXNetLite which is a distillation technique that is useful for transferring the knowledge from our ensemble model to an efficient deep neural network using label-smoothen soft labels for able to effectively detect expressions in real-time. Both of the techniques performed quite well, where the ensemble model (EmoXNet) helped to achieve 85.07% test accuracy on FER2013 with FER+ annotations and 86.25% test accuracy on RAF-DB. Moreover, the distilled model (EmoXNetLite) showed 82.07% test accuracy on FER2013 with FER+ annotations and 81.78% test accuracy on RAF-DB. Results show that our models seem to generalize well on new data and are learned to focus on relevant facial representations for expressions recognition.