English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep generative models have recently presented impressive results ingenerating realistic face images of random synthetic identities.  To generate multiple samples of a certain synthetic identity, previous worksproposed to disentangle the latent space of GANs by incorporating additionalsupervision or regularization, enabling the manipulation of certain attributes.Others proposed to disentangle specific factors in unconditional pretrainedGANs latent spaces to control their output, which also requires supervision byattribute classifiers. Moreover, these attributes are entangled in GAN's latentspace, making it difficult to manipulate them without affecting the identityinformation. We propose in this work a framework, ExFaceGAN, to disentangleidentity information in pretrained GANs latent spaces, enabling the generationof multiple samples of any synthetic identity. Given a reference latent code ofany synthetic image and latent space of pretrained GAN, our ExFaceGAN learns anidentity directional boundary that disentangles the latent space into twosub-spaces, with latent codes of samples that are either identity similar ordissimilar to a reference image. By sampling from each side of the boundary,our ExFaceGAN can generate multiple samples of synthetic identity without theneed for designing a dedicated architecture or supervision from attributeclassifiers. We demonstrate the generalizability and effectiveness of ExFaceGANby integrating it into learned latent spaces of three SOTA GAN approaches. Asan example of the practical benefit of our ExFaceGAN, we empirically prove thatdata generated by ExFaceGAN can be successfully used to train face recognitionmodels (\url{https://github.com/fdbtrs/ExFaceGAN}).

ExFaceGAN: Exploring Identity Directions in GAN's Learned Latent Space for Synthetic Identity Generation