Cargando…

SalfMix: A Novel Single Image-Based Data Augmentation Technique Using a Saliency Map

Modern data augmentation strategies such as Cutout, Mixup, and CutMix, have achieved good performance in image recognition tasks. Particularly, the data augmentation approaches, such as Mixup and CutMix, that mix two images to generate a mixed training image, could generalize convolutional neural ne...

Descripción completa

Detalles Bibliográficos
Autores principales: Choi, Jaehyeop, Lee, Chaehyeon, Lee, Donggyu, Jung, Heechul
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8706478/
https://www.ncbi.nlm.nih.gov/pubmed/34960539
http://dx.doi.org/10.3390/s21248444
_version_ 1784622202412335104
author Choi, Jaehyeop
Lee, Chaehyeon
Lee, Donggyu
Jung, Heechul
author_facet Choi, Jaehyeop
Lee, Chaehyeon
Lee, Donggyu
Jung, Heechul
author_sort Choi, Jaehyeop
collection PubMed
description Modern data augmentation strategies such as Cutout, Mixup, and CutMix, have achieved good performance in image recognition tasks. Particularly, the data augmentation approaches, such as Mixup and CutMix, that mix two images to generate a mixed training image, could generalize convolutional neural networks better than single image-based data augmentation approaches such as Cutout. We focus on the fact that the mixed image can improve generalization ability, and we wondered if it would be effective to apply it to a single image. Consequently, we propose a new data augmentation method to produce a self-mixed image based on a saliency map, called SalfMix. Furthermore, we combined SalfMix with state-of-the-art two images-based approaches, such as Mixup, SaliencyMix, and CutMix, to increase the performance, called HybridMix. The proposed SalfMix achieved better accuracies than Cutout, and HybridMix achieved state-of-the-art performance on three classification datasets: CIFAR-10, CIFAR-100, and TinyImageNet-200. Furthermore, HybridMix achieved the best accuracy in object detection tasks on the VOC dataset, in terms of mean average precision.
format Online
Article
Text
id pubmed-8706478
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-87064782021-12-25 SalfMix: A Novel Single Image-Based Data Augmentation Technique Using a Saliency Map Choi, Jaehyeop Lee, Chaehyeon Lee, Donggyu Jung, Heechul Sensors (Basel) Article Modern data augmentation strategies such as Cutout, Mixup, and CutMix, have achieved good performance in image recognition tasks. Particularly, the data augmentation approaches, such as Mixup and CutMix, that mix two images to generate a mixed training image, could generalize convolutional neural networks better than single image-based data augmentation approaches such as Cutout. We focus on the fact that the mixed image can improve generalization ability, and we wondered if it would be effective to apply it to a single image. Consequently, we propose a new data augmentation method to produce a self-mixed image based on a saliency map, called SalfMix. Furthermore, we combined SalfMix with state-of-the-art two images-based approaches, such as Mixup, SaliencyMix, and CutMix, to increase the performance, called HybridMix. The proposed SalfMix achieved better accuracies than Cutout, and HybridMix achieved state-of-the-art performance on three classification datasets: CIFAR-10, CIFAR-100, and TinyImageNet-200. Furthermore, HybridMix achieved the best accuracy in object detection tasks on the VOC dataset, in terms of mean average precision. MDPI 2021-12-17 /pmc/articles/PMC8706478/ /pubmed/34960539 http://dx.doi.org/10.3390/s21248444 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Choi, Jaehyeop
Lee, Chaehyeon
Lee, Donggyu
Jung, Heechul
SalfMix: A Novel Single Image-Based Data Augmentation Technique Using a Saliency Map
title SalfMix: A Novel Single Image-Based Data Augmentation Technique Using a Saliency Map
title_full SalfMix: A Novel Single Image-Based Data Augmentation Technique Using a Saliency Map
title_fullStr SalfMix: A Novel Single Image-Based Data Augmentation Technique Using a Saliency Map
title_full_unstemmed SalfMix: A Novel Single Image-Based Data Augmentation Technique Using a Saliency Map
title_short SalfMix: A Novel Single Image-Based Data Augmentation Technique Using a Saliency Map
title_sort salfmix: a novel single image-based data augmentation technique using a saliency map
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8706478/
https://www.ncbi.nlm.nih.gov/pubmed/34960539
http://dx.doi.org/10.3390/s21248444
work_keys_str_mv AT choijaehyeop salfmixanovelsingleimagebaseddataaugmentationtechniqueusingasaliencymap
AT leechaehyeon salfmixanovelsingleimagebaseddataaugmentationtechniqueusingasaliencymap
AT leedonggyu salfmixanovelsingleimagebaseddataaugmentationtechniqueusingasaliencymap
AT jungheechul salfmixanovelsingleimagebaseddataaugmentationtechniqueusingasaliencymap