Generating Dataset For Large-scale 3D Facial Emotion Recognition

TL;DR

This paper introduces a method to generate a large-scale 3D facial emotion dataset and trains a deep CNN on it, addressing data scarcity and advancing 3D FER performance.

Contribution

It presents a novel approach for creating extensive 3D facial datasets with emotion labels and develops a deep CNN trained on this data for improved 3D FER.

Findings

Achieved training on 624,000 3D facial scans

Tested on 208,000 3D facial scans

Enhanced accuracy in 3D facial emotion recognition

Abstract

The tremendous development in deep learning has led facial expression recognition (FER) to receive much attention in the past few years. Although 3D FER has an inherent edge over its 2D counterpart, work on 2D images has dominated the field. The main reason for the slow development of 3D FER is the unavailability of large training and large test datasets. Recognition accuracies have already saturated on existing 3D emotion recognition datasets due to their small gallery sizes. Unlike 2D photographs, 3D facial scans are not easy to collect, causing a bottleneck in the development of deep 3D FER networks and datasets. In this work, we propose a method for generating a large dataset of 3D faces with labeled emotions. We also develop a deep convolutional neural network(CNN) for 3D FER trained on 624,000 3D facial scans. The test data comprises 208,000 3D facial scans.