Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/33257
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | CHEN, Siang | - |
dc.contributor.author | Huang, Kai | - |
dc.contributor.author | LI, Bowen | - |
dc.contributor.author | Xiong, Dongliang | - |
dc.contributor.author | Jiang, Haitian | - |
dc.contributor.author | CLAESEN, Luc | - |
dc.date.accessioned | 2021-02-02T10:59:53Z | - |
dc.date.available | 2021-02-02T10:59:53Z | - |
dc.date.issued | 2021 | - |
dc.date.submitted | 2021-02-01T15:09:30Z | - |
dc.identifier.citation | Lecture notes in computer science, 12537, p. 119 -135 | - |
dc.identifier.isbn | 978-3-030-67069-6 | - |
dc.identifier.isbn | 978-3-030-67070-2 | - |
dc.identifier.issn | 0302-9743 | - |
dc.identifier.issn | 1611-3349 | - |
dc.identifier.uri | http://hdl.handle.net/1942/33257 | - |
dc.description.abstract | In recent years, remarkable progress has been made in single image super-resolution due to the powerful representation capabilities of deep neural networks. However, the superior performance is at the expense of excessive computation costs, limiting the SR application in resource-constrained devices. To address this problem, we firstly propose a hybrid composition block (HCB), which contains asymmetric and shrinked spatial convolution in parallel. Secondly, we build our baseline model based on cascaded HCB with a progressive upsampling method. Besides, feature fusion method is developed which concatenates all of the previous feature maps of HCB. Thirdly, to solve the misalignment problem in pruning residual networks, we propose a fine-grained channel pruning that allows adaptive connections to fully skip the residual block, and any unimportant channel between convolutions can be pruned independently. Finally, we present an adaptive hybrid composition based super-resolution network (AHCSRN) by pruning the baseline model. Extensive experiments demonstrate that the proposed method can achieve better performance than state-of-the-art SR models with ultra-low parameters and Flops. | - |
dc.description.sponsorship | This work is supported by the National Key R&D Program of China (2020YFB0906000, 2020YFB0906001). | - |
dc.language.iso | en | - |
dc.publisher | Springer, Cham | - |
dc.relation.ispartofseries | Lecture Notes in Computer Science | - |
dc.rights | Springer Nature Switzerland AG 2020 | - |
dc.subject.other | Single image super-resolution | - |
dc.subject.other | Efficient model | - |
dc.subject.other | Channel pruning | - |
dc.title | Adaptive Hybrid Composition Based Super-Resolution Network via Fine-Grained Channel Pruning | - |
dc.type | Journal Contribution | - |
local.bibliographicCitation.conferencedate | August 23–28, 2020 | - |
local.bibliographicCitation.conferencename | European Conference on Computer Vision | - |
local.bibliographicCitation.conferenceplace | Glasgow, UK | - |
dc.identifier.epage | 135 | - |
dc.identifier.spage | 119 | - |
dc.identifier.volume | 12537 | - |
local.bibliographicCitation.jcat | A1 | - |
local.type.refereed | Refereed | - |
local.type.specified | Article | - |
local.relation.ispartofseriesnr | 12537 | - |
local.type.programme | VSC | - |
dc.identifier.doi | 10.1007/978-3-030-67070-2_7 | - |
local.provider.type | CrossRef | - |
local.uhasselt.uhpub | yes | - |
local.uhasselt.international | yes | - |
item.fulltext | With Fulltext | - |
item.accessRights | Open Access | - |
item.fullcitation | CHEN, Siang; Huang, Kai; LI, Bowen; Xiong, Dongliang; Jiang, Haitian & CLAESEN, Luc (2021) Adaptive Hybrid Composition Based Super-Resolution Network via Fine-Grained Channel Pruning. In: Lecture notes in computer science, 12537, p. 119 -135. | - |
item.contributor | CHEN, Siang | - |
item.contributor | Huang, Kai | - |
item.contributor | LI, Bowen | - |
item.contributor | Xiong, Dongliang | - |
item.contributor | Jiang, Haitian | - |
item.contributor | CLAESEN, Luc | - |
crisitem.journal.issn | 0302-9743 | - |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Adaptive Hybrid Composition based Super-Resolution Network via Fine-grained Channel Pruning.pdf | Peer-reviewed author version | 4 MB | Adobe PDF | View/Open |
Page view(s)
60
checked on Sep 5, 2022
Download(s)
28
checked on Sep 5, 2022
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.