Detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network

Article Highlights We changed the classic network and loss function to obtain the global 3D depth information of the scene. A depth gradient acquisition scheme is designed to generate the local details of the scene. We can obtain a plausible depth map with better depth details through our developed...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Huihui Xu, Nan Liu
Formato:	article
Lenguaje:	EN
Publicado:	Springer 2021
Materias:	Depth prediction Fully convolutional residual network Gradient network Depth gradient Science Q Technology T
Acceso en línea:	https://doaj.org/article/f45e86bba2b9400f8c4fb691f02003f9
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:f45e86bba2b9400f8c4fb691f02003f9
record_format	dspace
spelling	oai:doaj.org-article:f45e86bba2b9400f8c4fb691f02003f92021-12-05T12:10:05ZDetail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network10.1007/s42452-021-04882-02523-39632523-3971https://doaj.org/article/f45e86bba2b9400f8c4fb691f02003f92021-11-01T00:00:00Zhttps://doi.org/10.1007/s42452-021-04882-0https://doaj.org/toc/2523-3963https://doaj.org/toc/2523-3971Article Highlights We changed the classic network and loss function to obtain the global 3D depth information of the scene. A depth gradient acquisition scheme is designed to generate the local details of the scene. We can obtain a plausible depth map with better depth details through our developed depth and depth gradients fusion strategy.Huihui XuNan LiuSpringerarticleDepth predictionFully convolutional residual networkGradient networkDepth gradientScienceQTechnologyTENSN Applied Sciences, Vol 3, Iss 12, Pp 1-15 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Depth prediction Fully convolutional residual network Gradient network Depth gradient Science Q Technology T
spellingShingle	Depth prediction Fully convolutional residual network Gradient network Depth gradient Science Q Technology T Huihui Xu Nan Liu Detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network
description	Article Highlights We changed the classic network and loss function to obtain the global 3D depth information of the scene. A depth gradient acquisition scheme is designed to generate the local details of the scene. We can obtain a plausible depth map with better depth details through our developed depth and depth gradients fusion strategy.
format	article
author	Huihui Xu Nan Liu
author_facet	Huihui Xu Nan Liu
author_sort	Huihui Xu
title	Detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network
title_short	Detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network
title_full	Detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network
title_fullStr	Detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network
title_full_unstemmed	Detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network
title_sort	detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network
publisher	Springer
publishDate	2021
url	https://doaj.org/article/f45e86bba2b9400f8c4fb691f02003f9
work_keys_str_mv	AT huihuixu detailpreservingdepthestimationfromasingleimagebasedonmodifiedfullyconvolutionalresidualnetworkandgradientnetwork AT nanliu detailpreservingdepthestimationfromasingleimagebasedonmodifiedfullyconvolutionalresidualnetworkandgradientnetwork
_version_	1718372228421124096

Detail-preserving depth estimation from a single image based on modified fully convolutional residual network and gradient network

Ejemplares similares