Pioneer Networks: Progressively Growing Generative Autoencoder

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Researchers

Research units

  • GenMind Ltd.

Abstract

We 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.

Details

Original languageEnglish
Title of host publicationComputer Vision – ACCV 2018 - 14th Asian Conference on Computer Vision, Revised Selected Papers
EditorsGreg Mori, C.V. Jawahar, Konrad Schindler, Hongdong Li
Publication statusPublished - 2019
MoE publication typeA4 Article in a conference publication
EventAsian Conference on Computer Vision - Perth, Australia
Duration: 2 Dec 20186 Dec 2018
Conference number: 14

Publication series

NameLecture notes in computer science
PublisherSpringer Nature
Volume11361
ISSN (Electronic)1611-3349

Conference

ConferenceAsian Conference on Computer Vision
Abbreviated titleACCV
CountryAustralia
CityPerth
Period02/12/201806/12/2018

    Research areas

  • Autoencoder, Computer vision, Generative models

Download statistics

No data available

ID: 33778897