An Approach for Improving Automatic Mouth Emotion Recognition

TL;DR

This paper presents a CNN-based method for automatic mouth emotion recognition aimed at aiding communication for individuals with health disorders, utilizing Haar classifiers for face and mouth detection in real-time applications.

Contribution

It introduces a generalized training approach for emotion recognition from mouth features using CNNs and Haar classifiers, expanding beyond personalized micro-expression systems.

Findings

Fast execution with Haar classifiers

Effective emotion recognition on generalized datasets

Supports real-time communication aid systems

Abstract

The study proposes and tests a technique for automated emotion recognition through mouth detection via Convolutional Neural Networks (CNN), meant to be applied for supporting people with health disorders with communication skills issues (e.g. muscle wasting, stroke, autism, or, more simply, pain) in order to recognize emotions and generate real-time feedback, or data feeding supporting systems. The software system starts the computation identifying if a face is present on the acquired image, then it looks for the mouth location and extracts the corresponding features. Both tasks are carried out using Haar Feature-based Classifiers, which guarantee fast execution and promising performance. If our previous works focused on visual micro-expressions for personalized training on a single user, this strategy aims to train the system also on generalized faces data sets.