Skip navigation
Skip navigation

Super-Resolving Very Low-Resolution Face Images with Supplementary Attributes

Yu, Xin; Fernando, Basura; Hartley, Richard; Porikli, Fatih

Description

Given a tiny face image, existing face hallucination methods aim at super-resolving its high-resolution (HR) counterpart by learning a mapping from an exemplar dataset. Since a low-resolution (LR) input patch may correspond to many HR candidate patches, this ambiguity may lead to distorted HR facial details and wrong attributes such as gender reversal. An LR input contains low-frequency facial components of its HR version while its residual face image, defined as the difference between the HR...[Show more]

dc.contributor.authorYu, Xin
dc.contributor.authorFernando, Basura
dc.contributor.authorHartley, Richard
dc.contributor.authorPorikli, Fatih
dc.date.accessioned2021-07-30T02:29:12Z
dc.identifier.isbn978-153866420-9
dc.identifier.issn1063-6919
dc.identifier.urihttp://hdl.handle.net/1885/241635
dc.description.abstractGiven a tiny face image, existing face hallucination methods aim at super-resolving its high-resolution (HR) counterpart by learning a mapping from an exemplar dataset. Since a low-resolution (LR) input patch may correspond to many HR candidate patches, this ambiguity may lead to distorted HR facial details and wrong attributes such as gender reversal. An LR input contains low-frequency facial components of its HR version while its residual face image, defined as the difference between the HR ground-truth and interpolated LR images, contains the missing high-frequency facial details. We demonstrate that supplementing residual images or feature maps with additional facial attribute information can significantly reduce the ambiguity in face super-resolution. To explore this idea, we develop an attribute-embedded upsampling network, which consists of an upsampling network and a discriminative network. The upsampling network is composed of an autoencoder with skip-connections, which incorporates facial attribute vectors into the residual features of LR inputs at the bottleneck of the autoencoder and deconvolutional layers used for upsampling. The discriminative network is designed to examine whether super-resolved faces contain the desired attributes or not and then its loss is used for updating the upsampling network. In this manner, we can super-resolve tiny (16×16 pixels) unaligned face images with a large upscaling factor of 8× while reducing the uncertainty of one-to-many mappings remarkably. By conducting extensive evaluations on a large-scale dataset, we demonstrate that our method achieves superior face hallucination results and outperforms the state-of-the-art.
dc.description.sponsorshipThis work was supported under the Australian Research Council‘s Discovery Projects funding scheme (project DP150104645)
dc.publisherIEEE
dc.relation.ispartofseries31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018
dc.rights© 2018 IEEE
dc.sourceIEEE/CVF Conference on Computer Vision and Pattern Recognition
dc.titleSuper-Resolving Very Low-Resolution Face Images with Supplementary Attributes
dc.typeJournal article
local.description.notesAdded manually as didn't import from ARIES
dc.date.issued2018-06-18
local.identifier.ariespublicationu3102795xPUB1515
local.publisher.urlhttps://ieeexplore.ieee.org/
local.type.statusPublished Version
local.contributor.affiliationYu, Xin, College of Engineering and Computer Science, ANU
local.contributor.affiliationFernando, Basura, College of Engineering and Computer Science, ANU
local.contributor.affiliationHartley, Richard, College of Engineering and Computer Science, ANU
local.contributor.affiliationPorikli, Fatih, College of Engineering and Computer Science, ANU
local.description.embargo2099-12-31
dc.relationhttp://purl.org/au-research/grants/arc/DP150104645
local.identifier.essn2575-7075
local.bibliographicCitation.startpage908
local.bibliographicCitation.lastpage917
local.identifier.doi10.1109/CVPR.2018.00101
CollectionsANU Research Publications

Download

File Description SizeFormat Image
Super-Resolving_Very_Low-Resolution_Face_Images_with_Supplementary_Attributes.pdf1.51 MBAdobe PDF    Request a copy


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

Updated:  19 May 2020/ Responsible Officer:  University Librarian/ Page Contact:  Library Systems & Web Coordinator