How far Generated Data Can Impact Neural Networks Performance?

Sayeh Gholipour Picha, Dawood Al Chanti, Alice Caplier

2023

Abstract

The success of deep learning models depends on the size and quality of the dataset to solve certain tasks. Here, we explore how far generated data can aid real data in improving the performance of Neural Networks. In this work, we consider facial expression recognition since it requires challenging local data generation at the level of local regions such as mouth, eyebrows, etc, rather than simple augmentation. Generative Adversarial Networks (GANs) provide an alternative method for generating such local deformations but they need further validation. To answer our question, we consider noncomplex Convolutional Neural Networks (CNNs) based classifiers for recognizing Ekman emotions. For the data generation process, we consider generating facial expressions (FEs) by relying on two GANs. The first generates a random identity while the second imposes facial deformations on top of it. We consider training the CNN classifier using FEs from: real-faces, GANs-generated, and finally using a combination of real and GAN-generated faces. We determine an upper bound regarding the data generation quantity to be mixed with the real one which contributes the most to enhancing FER accuracy. In our experiments, we find out that 5-times more synthetic data to the real FEs dataset increases accuracy by 16%.

Paper Citation

in Harvard Style

Gholipour Picha S., Al Chanti D. and Caplier A. (2023). How far Generated Data Can Impact Neural Networks Performance?. In Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP; ISBN 978-989-758-634-7, SciTePress, pages 472-479. DOI: 10.5220/0011629000003417

in Bibtex Style

@conference{visapp23,
author={Sayeh Gholipour Picha and Dawood Al Chanti and Alice Caplier},
title={How far Generated Data Can Impact Neural Networks Performance?},
booktitle={Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP},
year={2023},
pages={472-479},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011629000003417},
isbn={978-989-758-634-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP
TI - How far Generated Data Can Impact Neural Networks Performance?
SN - 978-989-758-634-7
AU - Gholipour Picha S.
AU - Al Chanti D.
AU - Caplier A.
PY - 2023
SP - 472
EP - 479
DO - 10.5220/0011629000003417
PB - SciTePress