Skip navigation
Skip navigation

Stereo Super-Resolution via a Deep Convolutional Network

Li, Junxuan; You, Shaodi; Robles-Kelly, Antonio

Description

In this paper, we present a method for stereo super-resolution which employs a deep network. The network is trained using the residual image so as to obtain a high resolution image from two, low resolution views. Our network is comprised by two deep sub-nets which share, at their output, a single convolutional layer. This last layer in the network delivers an estimate of the residual image which is then used, in combination with the left input frame of the stereo pair, to compute the...[Show more]

dc.contributor.authorLi, Junxuan
dc.contributor.authorYou, Shaodi
dc.contributor.authorRobles-Kelly, Antonio
dc.contributor.editorGuo, Y.
dc.contributor.editorMurshed, M.
dc.contributor.editorWang, Z.
dc.contributor.editorFeng, D.
dc.contributor.editorLi, H.
dc.contributor.editorCai, W.
dc.contributor.editorGao, J.
dc.coverage.spatialSydney, Australia
dc.date.accessioned2020-06-24T01:42:39Z
dc.date.createdNovember 29 - December 1 2017
dc.identifier.isbn9781538628393
dc.identifier.urihttp://hdl.handle.net/1885/205482
dc.description.abstractIn this paper, we present a method for stereo super-resolution which employs a deep network. The network is trained using the residual image so as to obtain a high resolution image from two, low resolution views. Our network is comprised by two deep sub-nets which share, at their output, a single convolutional layer. This last layer in the network delivers an estimate of the residual image which is then used, in combination with the left input frame of the stereo pair, to compute the super-resolved image at output. Each of these sub- networks is comprised by ten weight layers and, hence, allows our network to combine structural information in the image across image regions efficiently. Moreover, by learning the residual image, the network copes better with vanishing gradients and its devoid of gradient clipping operations. We illustrate the utility of our network for image-pair super-resolution and compare our network to its non-gradient trained analogue and alternatives elsewhere in the literature.
dc.format.mimetypeapplication/pdf
dc.language.isoen_AU
dc.publisherIEEE
dc.relation.ispartofseries2017 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2017
dc.rights© 2017 IEEE
dc.sourceDICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications
dc.titleStereo Super-Resolution via a Deep Convolutional Network
dc.typeConference paper
local.description.notesImported from ARIES
local.description.refereedYes
dc.date.issued2017
local.identifier.absfor080104 - Computer Vision
local.identifier.ariespublicationu4485658xPUB502
local.publisher.urlhttps://www.ieee.org/
local.type.statusPublished Version
local.contributor.affiliationLi, Junxuan, College of Engineering and Computer Science, ANU
local.contributor.affiliationYou, Shaodi, College of Engineering and Computer Science, ANU
local.contributor.affiliationRobles-Kelly, Antonio, College of Engineering and Computer Science, ANU
local.description.embargo2037-12-31
local.bibliographicCitation.startpage1
local.bibliographicCitation.lastpage7
local.identifier.doi10.1109/DICTA.2017.8227492
dc.date.updated2020-01-19T07:32:16Z
local.identifier.scopusID2-s2.0-85048331631
CollectionsANU Research Publications

Download

File Description SizeFormat Image
01_Li_Stereo_Super-Resolution_via_a_2017.pdf5.15 MBAdobe PDF    Request a copy


Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.

Updated:  17 November 2022/ Responsible Officer:  University Librarian/ Page Contact:  Library Systems & Web Coordinator