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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In recent years, Generative Adversarial Networks have become ubiquitous inboth research and public perception, but how GANs convert an unstructuredlatent code to a high quality output is still an open question. In this work,we investigate regression into the latent space as a probe to understand thecompositional properties of GANs. We find that combining the regressor and apretrained generator provides a strong image prior, allowing us to createcomposite images from a collage of random image parts at inference time whilemaintaining global consistency. To compare compositional properties acrossdifferent generators, we measure the trade-offs between reconstruction of theunrealistic input and image quality of the regenerated samples. We find thatthe regression approach enables more localized editing of individual imageparts compared to direct editing in the latent space, and we conductexperiments to quantify this independence effect. Our method is agnostic to thesemantics of edits, and does not require labels or predefined concepts duringtraining. Beyond image composition, our method extends to a number of relatedapplications, such as image inpainting or example-based image editing, which wedemonstrate on several GANs and datasets, and because it uses only a singleforward pass, it can operate in real-time. Code is available on our projectpage: https://chail.github.io/latent-composition/.

Using latent space regression to analyze and leverage compositionality in GANs