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

The image-to-image translation (I2IT) model takes a target label or areference image as the input, and changes a source into the specified targetdomain. The two types of synthesis, either label- or reference-based, havesubstantial differences. Particularly, the label-based synthesis reflects thecommon characteristics of the target domain, and the reference-based shows thespecific style similar to the reference. This paper intends to bridge the gapbetween them in the task of multi-attribute I2IT. We design the label- andreference-based encoding modules (LEM and REM) to compare the domaindifferences. They first transfer the source image and target label (orreference) into a common embedding space, by providing the opposite directionsthrough the attribute difference vector. Then the two embeddings are simplyfused together to form the latent code S_rand (or S_ref), reflecting the domainstyle differences, which is injected into each layer of the generator by SPADE.To link LEM and REM, so that two types of results benefit each other, weencourage the two latent codes to be close, and set up the cycle consistencybetween the forward and backward translations on them. Moreover, theinterpolation between the S_rand and S_ref is also used to synthesize an extraimage. Experiments show that label- and reference-based synthesis are indeedmutually promoted, so that we can have the diverse results from LEM, and highquality results with the similar style of the reference.

Bridging the Gap between Label- and Reference-based Synthesis in Multi-attribute Image-to-Image Translation