Deep Convolutional Neural Networks for Smile Recognition

TL;DR

This thesis develops a deep convolutional neural network-based smile detector that achieves over 99% accuracy on a standard database, significantly surpassing previous methods, and demonstrates the effectiveness of GPU acceleration.

Contribution

It introduces a novel deep CNN model for smile recognition, optimized with comprehensive database analysis and GPU training, achieving state-of-the-art accuracy.

Findings

Achieved 99.45% accuracy on DISFA database

Outperformed existing methods with 65.55% to 79.67% accuracy range

GPU implementation provided up to 10x speedup

Abstract

This thesis describes the design and implementation of a smile detector based on deep convolutional neural networks. It starts with a summary of neural networks, the difficulties of training them and new training methods, such as Restricted Boltzmann Machines or autoencoders. It then provides a literature review of convolutional neural networks and recurrent neural networks. In order to select databases for smile recognition, comprehensive statistics of databases popular in the field of facial expression recognition were generated and are summarized in this thesis. It then proposes a model for smile detection, of which the main part is implemented. The experimental results are discussed in this thesis and justified based on a comprehensive model selection performed. All experiments were run on a Tesla K40c GPU benefiting from a speedup of up to factor 10 over the computations on a CPU. A smile detection test accuracy of 99.45% is achieved for the Denver Intensity of Spontaneous Facial Action (DISFA) database, significantly outperforming existing approaches with accuracies ranging from 65.55% to 79.67%. This experiment is re-run under various variations, such as retaining less neutral images or only the low or high intensities, of which the results are extensively compared.