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

Variational Autoencoders (VAE) are probabilistic deep generative modelsunderpinned by elegant theory, stable training processes, and meaningfulmanifold representations. However, they produce blurry images due to a lack ofexplicit emphasis over high-frequency textural details of the images, and thedifficulty to directly model the complex joint probability distribution overthe high-dimensional image space. In this work, we approach these twochallenges with a novel wavelet space VAE that uses the decoder to model theimages in the wavelet coefficient space. This enables the VAE to emphasize overhigh-frequency components within an image obtained via wavelet decomposition.Additionally, by decomposing the complex function of generatinghigh-dimensional images into inverse wavelet transformation and generation ofwavelet coefficients, the latter becomes simpler to model by the VAE. Weempirically validate that deep generative models operating in the wavelet spacecan generate images of higher quality than the image (RGB) space counterparts.Quantitatively, on benchmark natural image datasets, we achieve consistentlybetter FID scores than VAE based architectures and competitive FID scores witha variety of GAN models for the same architectural and experimental setup.Furthermore, the proposed wavelet-based generative model retains desirableattributes like disentangled and informative latent representation withoutlosing the quality in the generated samples.

Wavelets to the Rescue: Improving Sample Quality of Latent Variable Deep Generative Models