Publication: Self-organized variational autoencoders (self-vae) for learned image compression
dc.contributor.coauthor | Malik, J. | |
dc.contributor.coauthor | Kıranyaz S. | |
dc.contributor.department | Department of Electrical and Electronics Engineering | |
dc.contributor.department | Graduate School of Sciences and Engineering | |
dc.contributor.kuauthor | Güven, Hilal | |
dc.contributor.kuauthor | Keleş, Onur | |
dc.contributor.kuauthor | Tekalp, Ahmet Murat | |
dc.contributor.kuauthor | Yılmaz, Mustafa Akın | |
dc.contributor.schoolcollegeinstitute | College of Engineering | |
dc.contributor.schoolcollegeinstitute | GRADUATE SCHOOL OF SCIENCES AND ENGINEERING | |
dc.date.accessioned | 2024-11-09T13:50:24Z | |
dc.date.issued | 2021 | |
dc.description.abstract | In end-to-end optimized learned image compression, it is standard practice to use a convolutional variational autoencoder with generalized divisive normalization (GDN) to transform images into a latent space. Recently, Operational Neural Networks (ONNs) that learn the best non-linearity from a set of alternatives, and their “self-organized” variants, Self-ONNs, that approximate any non-linearity via Taylor series have been proposed to address the limitations of convolutional layers and a fixed nonlinear activation. In this paper, we propose to replace the convolutional and GDN layers in the variational autoencoder with self-organized operational layers, and propose a novel self-organized variational autoencoder (Self-VAE) architecture that benefits from stronger non-linearity. The experimental results demonstrate that the proposed Self-VAE yields improvements in both rate-distortion performance and perceptual image quality. | |
dc.description.fulltext | YES | |
dc.description.indexedby | WOS | |
dc.description.indexedby | Scopus | |
dc.description.openaccess | YES | |
dc.description.publisherscope | International | |
dc.description.sponsoredbyTubitakEu | TÜBİTAK | |
dc.description.sponsorship | Scientific and Technological Research Council of Turkey (TÜBİTAK) | |
dc.description.sponsorship | Koç Üniversitesi İş Bankası Yapay Zeka Uygulama ve Araştırma Merkezi (KUIS AI)/ Koç University İş Bank Artificial Intelligence Center (KUIS AI) | |
dc.description.sponsorship | Turkish Academy of Sciences (TÜBA) | |
dc.description.version | Publisher version | |
dc.identifier.doi | 10.1109/ICIP42928.2021.9506041 | |
dc.identifier.embargo | NO | |
dc.identifier.filenameinventoryno | IR03568 | |
dc.identifier.isbn | 9.78167E+12 | |
dc.identifier.issn | 15224880 | |
dc.identifier.quartile | N/A | |
dc.identifier.scopus | 2-s2.0-85125590801 | |
dc.identifier.uri | https://hdl.handle.net/20.500.14288/3908 | |
dc.identifier.wos | 819455103170 | |
dc.keywords | End-to-end learned image compression | |
dc.keywords | Perceptual quality metrics | |
dc.keywords | Rate-distortion performance | |
dc.keywords | Self-organized operational layer | |
dc.keywords | Variational autoencoder | |
dc.language.iso | eng | |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) | |
dc.relation.grantno | 120C156, 217E033 | |
dc.relation.ispartof | Proceedings - International Conference on Image Processing, ICIP | |
dc.relation.uri | http://cdm21054.contentdm.oclc.org/cdm/ref/collection/IR/id/10434 | |
dc.subject | Compression | |
dc.subject | JPEG | |
dc.subject | Deep learning | |
dc.title | Self-organized variational autoencoders (self-vae) for learned image compression | |
dc.type | Conference Proceeding | |
dspace.entity.type | Publication | |
local.contributor.kuauthor | Tekalp, Ahmet Murat | |
local.contributor.kuauthor | Keleş, Onur | |
local.contributor.kuauthor | Yılmaz, Mustafa Akın | |
local.contributor.kuauthor | Güven, Hilal | |
local.publication.orgunit1 | College of Engineering | |
local.publication.orgunit1 | GRADUATE SCHOOL OF SCIENCES AND ENGINEERING | |
local.publication.orgunit2 | Department of Electrical and Electronics Engineering | |
local.publication.orgunit2 | Graduate School of Sciences and Engineering | |
relation.isOrgUnitOfPublication | 21598063-a7c5-420d-91ba-0cc9b2db0ea0 | |
relation.isOrgUnitOfPublication | 3fc31c89-e803-4eb1-af6b-6258bc42c3d8 | |
relation.isOrgUnitOfPublication.latestForDiscovery | 21598063-a7c5-420d-91ba-0cc9b2db0ea0 | |
relation.isParentOrgUnitOfPublication | 8e756b23-2d4a-4ce8-b1b3-62c794a8c164 | |
relation.isParentOrgUnitOfPublication | 434c9663-2b11-4e66-9399-c863e2ebae43 | |
relation.isParentOrgUnitOfPublication.latestForDiscovery | 8e756b23-2d4a-4ce8-b1b3-62c794a8c164 |
Files
Original bundle
1 - 1 of 1