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

Recent advances in deep learning have witnessed many successful unsupervisedimage-to-image translation models that learn correspondences between two visualdomains without paired data. However, it is still a great challenge to buildrobust mappings between various domains especially for those with drasticvisual discrepancies. In this paper, we introduce a novel versatile framework,Generative Prior-guided UNsupervised Image-to-image Translation (GP-UNIT), thatimproves the quality, applicability and controllability of the existingtranslation models. The key idea of GP-UNIT is to distill the generative priorfrom pre-trained class-conditional GANs to build coarse-level cross-domaincorrespondences, and to apply the learned prior to adversarial translations toexcavate fine-level correspondences. With the learned multi-level contentcorrespondences, GP-UNIT is able to perform valid translations between bothclose domains and distant domains. For close domains, GP-UNIT can beconditioned on a parameter to determine the intensity of the contentcorrespondences during translation, allowing users to balance between contentand style consistency. For distant domains, semi-supervised learning isexplored to guide GP-UNIT to discover accurate semantic correspondences thatare hard to learn solely from the appearance. We validate the superiority ofGP-UNIT over state-of-the-art translation models in robust, high-quality anddiversified translations between various domains through extensive experiments.

GP-UNIT: Generative Prior for Versatile Unsupervised Image-to-Image Translation