Cargando…

Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model

In this paper, we propose a monocular catadioptric panoramic depth estimation algorithm based on an improved end-to-end neural network model. First, we use an enhanced concentric circle approximation unfolding algorithm to unfold the panoramic images captured by the catadioptric panoramic camera and...

Descripción completa

Detalles Bibliográficos
Autores principales: Yan, Fei, Liu, Lan, Ding, Xupeng, Zhang, Qiong, Liu, Yunqing
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10560983/
https://www.ncbi.nlm.nih.gov/pubmed/37818233
http://dx.doi.org/10.3389/fnbot.2023.1278986
_version_ 1785117825453522944
author Yan, Fei
Liu, Lan
Ding, Xupeng
Zhang, Qiong
Liu, Yunqing
author_facet Yan, Fei
Liu, Lan
Ding, Xupeng
Zhang, Qiong
Liu, Yunqing
author_sort Yan, Fei
collection PubMed
description In this paper, we propose a monocular catadioptric panoramic depth estimation algorithm based on an improved end-to-end neural network model. First, we use an enhanced concentric circle approximation unfolding algorithm to unfold the panoramic images captured by the catadioptric panoramic camera and then extract the effective regions. In addition, the integration of the Non-local attention mechanism is exploited to improve image understanding. Finally, a depth smoothness loss strategy is implemented to further enhance the reliability and precision of the estimated depths. Experimental results confirm that this refined algorithm is capable of providing highly accurate estimates of object depth.
format Online
Article
Text
id pubmed-10560983
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-105609832023-10-10 Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model Yan, Fei Liu, Lan Ding, Xupeng Zhang, Qiong Liu, Yunqing Front Neurorobot Neuroscience In this paper, we propose a monocular catadioptric panoramic depth estimation algorithm based on an improved end-to-end neural network model. First, we use an enhanced concentric circle approximation unfolding algorithm to unfold the panoramic images captured by the catadioptric panoramic camera and then extract the effective regions. In addition, the integration of the Non-local attention mechanism is exploited to improve image understanding. Finally, a depth smoothness loss strategy is implemented to further enhance the reliability and precision of the estimated depths. Experimental results confirm that this refined algorithm is capable of providing highly accurate estimates of object depth. Frontiers Media S.A. 2023-09-25 /pmc/articles/PMC10560983/ /pubmed/37818233 http://dx.doi.org/10.3389/fnbot.2023.1278986 Text en Copyright © 2023 Yan, Liu, Ding, Zhang and Liu. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neuroscience
Yan, Fei
Liu, Lan
Ding, Xupeng
Zhang, Qiong
Liu, Yunqing
Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model
title Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model
title_full Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model
title_fullStr Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model
title_full_unstemmed Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model
title_short Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model
title_sort monocular catadioptric panoramic depth estimation via improved end-to-end neural network model
topic Neuroscience
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10560983/
https://www.ncbi.nlm.nih.gov/pubmed/37818233
http://dx.doi.org/10.3389/fnbot.2023.1278986
work_keys_str_mv AT yanfei monocularcatadioptricpanoramicdepthestimationviaimprovedendtoendneuralnetworkmodel
AT liulan monocularcatadioptricpanoramicdepthestimationviaimprovedendtoendneuralnetworkmodel
AT dingxupeng monocularcatadioptricpanoramicdepthestimationviaimprovedendtoendneuralnetworkmodel
AT zhangqiong monocularcatadioptricpanoramicdepthestimationviaimprovedendtoendneuralnetworkmodel
AT liuyunqing monocularcatadioptricpanoramicdepthestimationviaimprovedendtoendneuralnetworkmodel