Cargando…

Agrast-6: Abridged VGG-Based Reflected Lightweight Architecture for Binary Segmentation of Depth Images Captured by Kinect

Binary object segmentation is a sub-area of semantic segmentation that could be used for a variety of applications. Semantic segmentation models could be applied to solve binary segmentation problems by introducing only two classes, but the models to solve this problem are more complex than actually...

Descripción completa

Detalles Bibliográficos
Autores principales: Ryselis, Karolis, Blažauskas, Tomas, Damaševičius, Robertas, Maskeliūnas, Rytis
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9460068/
https://www.ncbi.nlm.nih.gov/pubmed/36080813
http://dx.doi.org/10.3390/s22176354
_version_ 1784786660741873664
author Ryselis, Karolis
Blažauskas, Tomas
Damaševičius, Robertas
Maskeliūnas, Rytis
author_facet Ryselis, Karolis
Blažauskas, Tomas
Damaševičius, Robertas
Maskeliūnas, Rytis
author_sort Ryselis, Karolis
collection PubMed
description Binary object segmentation is a sub-area of semantic segmentation that could be used for a variety of applications. Semantic segmentation models could be applied to solve binary segmentation problems by introducing only two classes, but the models to solve this problem are more complex than actually required. This leads to very long training times, since there are usually tens of millions of parameters to learn in this category of convolutional neural networks (CNNs). This article introduces a novel abridged VGG-16 and SegNet-inspired reflected architecture adapted for binary segmentation tasks. The architecture has 27 times fewer parameters than SegNet but yields 86% segmentation cross-intersection accuracy and 93% binary accuracy. The proposed architecture is evaluated on a large dataset of depth images collected using the Kinect device, achieving an accuracy of 99.25% in human body shape segmentation and 87% in gender recognition tasks.
format Online
Article
Text
id pubmed-9460068
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-94600682022-09-10 Agrast-6: Abridged VGG-Based Reflected Lightweight Architecture for Binary Segmentation of Depth Images Captured by Kinect Ryselis, Karolis Blažauskas, Tomas Damaševičius, Robertas Maskeliūnas, Rytis Sensors (Basel) Article Binary object segmentation is a sub-area of semantic segmentation that could be used for a variety of applications. Semantic segmentation models could be applied to solve binary segmentation problems by introducing only two classes, but the models to solve this problem are more complex than actually required. This leads to very long training times, since there are usually tens of millions of parameters to learn in this category of convolutional neural networks (CNNs). This article introduces a novel abridged VGG-16 and SegNet-inspired reflected architecture adapted for binary segmentation tasks. The architecture has 27 times fewer parameters than SegNet but yields 86% segmentation cross-intersection accuracy and 93% binary accuracy. The proposed architecture is evaluated on a large dataset of depth images collected using the Kinect device, achieving an accuracy of 99.25% in human body shape segmentation and 87% in gender recognition tasks. MDPI 2022-08-24 /pmc/articles/PMC9460068/ /pubmed/36080813 http://dx.doi.org/10.3390/s22176354 Text en © 2022 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
Ryselis, Karolis
Blažauskas, Tomas
Damaševičius, Robertas
Maskeliūnas, Rytis
Agrast-6: Abridged VGG-Based Reflected Lightweight Architecture for Binary Segmentation of Depth Images Captured by Kinect
title Agrast-6: Abridged VGG-Based Reflected Lightweight Architecture for Binary Segmentation of Depth Images Captured by Kinect
title_full Agrast-6: Abridged VGG-Based Reflected Lightweight Architecture for Binary Segmentation of Depth Images Captured by Kinect
title_fullStr Agrast-6: Abridged VGG-Based Reflected Lightweight Architecture for Binary Segmentation of Depth Images Captured by Kinect
title_full_unstemmed Agrast-6: Abridged VGG-Based Reflected Lightweight Architecture for Binary Segmentation of Depth Images Captured by Kinect
title_short Agrast-6: Abridged VGG-Based Reflected Lightweight Architecture for Binary Segmentation of Depth Images Captured by Kinect
title_sort agrast-6: abridged vgg-based reflected lightweight architecture for binary segmentation of depth images captured by kinect
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9460068/
https://www.ncbi.nlm.nih.gov/pubmed/36080813
http://dx.doi.org/10.3390/s22176354
work_keys_str_mv AT ryseliskarolis agrast6abridgedvggbasedreflectedlightweightarchitectureforbinarysegmentationofdepthimagescapturedbykinect
AT blazauskastomas agrast6abridgedvggbasedreflectedlightweightarchitectureforbinarysegmentationofdepthimagescapturedbykinect
AT damaseviciusrobertas agrast6abridgedvggbasedreflectedlightweightarchitectureforbinarysegmentationofdepthimagescapturedbykinect
AT maskeliunasrytis agrast6abridgedvggbasedreflectedlightweightarchitectureforbinarysegmentationofdepthimagescapturedbykinect