Cargando…

Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model

Extended from superpixel segmentation by adding an additional constraint on temporal consistency, supervoxel segmentation is to partition video frames into atomic segments. In this work, we propose a novel scheme for supervoxel segmentation to address the problem of new and moving objects, where the...

Descripción completa

Detalles Bibliográficos
Autores principales: Ban, Zhihua, Chen, Zhong, Liu, Jianguo
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2018
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5795368/
https://www.ncbi.nlm.nih.gov/pubmed/29303972
http://dx.doi.org/10.3390/s18010128
_version_ 1783297282153644032
author Ban, Zhihua
Chen, Zhong
Liu, Jianguo
author_facet Ban, Zhihua
Chen, Zhong
Liu, Jianguo
author_sort Ban, Zhihua
collection PubMed
description Extended from superpixel segmentation by adding an additional constraint on temporal consistency, supervoxel segmentation is to partition video frames into atomic segments. In this work, we propose a novel scheme for supervoxel segmentation to address the problem of new and moving objects, where the segmentation is performed on every two consecutive frames and thus each internal frame has two valid superpixel segmentations. This scheme provides coarse-grained parallel ability, and subsequent algorithms can validate their result using two segmentations that will further improve robustness. To implement this scheme, a voxel-related Gaussian mixture model (GMM) is proposed, in which each supervoxel is assumed to be distributed in a local region and represented by two Gaussian distributions that share the same color parameters to capture temporal consistency. Our algorithm has a lower complexity with respect to frame size than the traditional GMM. According to our experiments, it also outperforms the state-of-the-art in accuracy.
format Online
Article
Text
id pubmed-5795368
institution National Center for Biotechnology Information
language English
publishDate 2018
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-57953682018-02-13 Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model Ban, Zhihua Chen, Zhong Liu, Jianguo Sensors (Basel) Article Extended from superpixel segmentation by adding an additional constraint on temporal consistency, supervoxel segmentation is to partition video frames into atomic segments. In this work, we propose a novel scheme for supervoxel segmentation to address the problem of new and moving objects, where the segmentation is performed on every two consecutive frames and thus each internal frame has two valid superpixel segmentations. This scheme provides coarse-grained parallel ability, and subsequent algorithms can validate their result using two segmentations that will further improve robustness. To implement this scheme, a voxel-related Gaussian mixture model (GMM) is proposed, in which each supervoxel is assumed to be distributed in a local region and represented by two Gaussian distributions that share the same color parameters to capture temporal consistency. Our algorithm has a lower complexity with respect to frame size than the traditional GMM. According to our experiments, it also outperforms the state-of-the-art in accuracy. MDPI 2018-01-05 /pmc/articles/PMC5795368/ /pubmed/29303972 http://dx.doi.org/10.3390/s18010128 Text en © 2018 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 (http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) ).
spellingShingle Article
Ban, Zhihua
Chen, Zhong
Liu, Jianguo
Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model
title Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model
title_full Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model
title_fullStr Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model
title_full_unstemmed Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model
title_short Supervoxel Segmentation with Voxel-Related Gaussian Mixture Model
title_sort supervoxel segmentation with voxel-related gaussian mixture model
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5795368/
https://www.ncbi.nlm.nih.gov/pubmed/29303972
http://dx.doi.org/10.3390/s18010128
work_keys_str_mv AT banzhihua supervoxelsegmentationwithvoxelrelatedgaussianmixturemodel
AT chenzhong supervoxelsegmentationwithvoxelrelatedgaussianmixturemodel
AT liujianguo supervoxelsegmentationwithvoxelrelatedgaussianmixturemodel