Pioneer Networks

dc.contributorAalto-yliopistofi
dc.contributorAalto Universityen
dc.contributor.authorHeljakka, Arien_US
dc.contributor.authorSolin, Arnoen_US
dc.contributor.authorKannala, Juhoen_US
dc.contributor.departmentDepartment of Computer Scienceen_US
dc.contributor.editorMori, Gregen_US
dc.contributor.editorJawahar, C.V.en_US
dc.contributor.editorSchindler, Konraden_US
dc.contributor.editorLi, Hongdongen_US
dc.date.accessioned2019-06-20T13:14:01Z
dc.date.available2019-06-20T13:14:01Z
dc.date.issued2019en_US
dc.description.abstractWe introduce a novel generative autoencoder network model that learns to encode and reconstruct images with high quality and resolution, and supports smooth random sampling from the latent space of the encoder. Generative adversarial networks (GANs) are known for their ability to simulate random high-quality images, but they cannot reconstruct existing images. Previous works have attempted to extend GANs to support such inference but, so far, have not delivered satisfactory high-quality results. Instead, we propose the Progressively Growing Generative Autoencoder (Pioneer) network which achieves high-quality reconstruction with images without requiring a GAN discriminator. We merge recent techniques for progressively building up the parts of the network with the recently introduced adversarial encoder–generator network. The ability to reconstruct input images is crucial in many real-world applications, and allows for precise intelligent manipulation of existing images. We show promising results in image synthesis and inference, with state-of-the-art results in CelebA inference tasks.en
dc.description.versionPeer revieweden
dc.format.extent22-38
dc.format.mimetypeapplication/pdfen_US
dc.identifier.citationHeljakka , A , Solin , A & Kannala , J 2019 , Pioneer Networks : Progressively Growing Generative Autoencoder . in G Mori , C V Jawahar , K Schindler & H Li (eds) , Computer Vision – ACCV 2018 - 14th Asian Conference on Computer Vision, Revised Selected Papers . Lecture notes in computer science , vol. 11361 , Springer , Springer, Cham , pp. 22-38 , Asian Conference on Computer Vision , Perth , Western Australia , Australia , 02/12/2018 . https://doi.org/10.1007/978-3-030-20887-5_2en
dc.identifier.doi10.1007/978-3-030-20887-5_2en_US
dc.identifier.isbn978-3-030-20886-8
dc.identifier.isbnOnline ISBN 978-3-030-20887-5
dc.identifier.issn1611-3349
dc.identifier.otherPURE UUID: 68b6dc32-099f-49d4-996f-f47af6001c0een_US
dc.identifier.otherPURE ITEMURL: https://research.aalto.fi/en/publications/68b6dc32-099f-49d4-996f-f47af6001c0een_US
dc.identifier.otherPURE LINK: http://www.scopus.com/inward/record.url?scp=85066805190&partnerID=8YFLogxKen_US
dc.identifier.otherPURE LINK: http://arxiv.org/abs/1807.03026en_US
dc.identifier.otherPURE FILEURL: https://research.aalto.fi/files/33778899/SCI_Heljakka_et_al_Pioneer_Networks_ACCV.pdfen_US
dc.identifier.urihttps://aaltodoc.aalto.fi/handle/123456789/38827
dc.identifier.urnURN:NBN:fi:aalto-201906203893
dc.language.isoenen
dc.publisherSpringer Nature
dc.relation.ispartofAsian Conference on Computer Visionen
dc.relation.ispartofseriesComputer Vision – ACCV 2018en
dc.relation.ispartofseriesLecture notes in computer scienceen
dc.relation.ispartofseriesVolume 11361en
dc.rightsopenAccessen
dc.subject.keywordAutoencoderen_US
dc.subject.keywordComputer visionen_US
dc.subject.keywordGenerative modelsen_US
dc.titlePioneer Networksen
dc.typeConference article in proceedingsfi
dc.type.versionacceptedVersion
Files