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 (VAEs) are powerful generative models, however theirgenerated samples are known to suffer from a characteristic blurriness, ascompared to the outputs of alternative generating techniques. Extensiveresearch efforts have been made to tackle this problem, and several works havefocused on modifying the reconstruction term of the evidence lower bound(ELBO). In particular, many have experimented with augmenting thereconstruction loss with losses in the frequency domain. Such loss functionsusually employ the Fourier transform to explicitly penalise the lack of higherfrequency components in the generated samples, which are responsible for sharpvisual features. In this paper, we explore the aspects of previous suchapproaches which aren't well understood, and we propose an augmentation to thereconstruction term in response to them. Our reasoning leads us to use theshort-time Fourier transform and to emphasise on local phase coherence betweenthe input and output samples. We illustrate the potential of our proposed losson the MNIST dataset by providing both qualitative and quantitative results.

Short-Time Fourier Transform for deblurring Variational Autoencoders