Variational autoencoder for image-based augmentation of eye-tracking data

Elbattah, Mahmoud; Loughnane, Colm; Gu�rin, Jean Luc; Carette, Romuald; Cilia, Federica; Dequen, Gilles

Abstract

Over the past decade, deep learning has achieved unprecedented successes in a diversity of application domains, given large-scale datasets. However, particular domains, such as healthcare, inherently suffer from data paucity and imbalance. Moreover, datasets could be largely inaccessible due to privacy concerns, or lack of data-sharing incentives. Such challenges have attached significance to the application of generative modeling and data augmentation in that domain. In this context, this study explores a machine learning-based approach for generating synthetic eye-tracking data. We explore a novel application of variational autoencoders (VAEs) in this regard. More specifically, a VAE model is trained to generate an image-based representation of the eye-tracking output, so-called scanpaths. Overall, our results validate that the VAE model could generate a plausible output from a limited dataset. Finally, it is empirically demonstrated that such approach could be employed as a mechanism for data augmentation to improve the performance in classification tasks.

• Journal Article Type: Article
• Acceptance Date: May 1, 2021
• Online Publication Date: May 3, 2021
• Publication Date: May 1, 2021
• Deposit Date: Apr 26, 2022
• Publicly Available Date: Apr 27, 2022
• Journal: Journal of Imaging
• Electronic ISSN: 2313-433X
• Publisher: MDPI
• Peer Reviewed: Peer Reviewed
• Volume: 7
• Issue: 5
• Article Number: 83
• DOI: https://doi.org/10.3390/jimaging7050083
• Keywords: Electrical and Electronic Engineering; Computer Graphics and Computer-Aided Design; Computer Vision and Pattern Recognition; Radiology, Nuclear Medicine and imaging
• Public URL: https://uwe-repository.worktribe.com/output/9187332