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

The Diffusion Model (DM) has emerged as the SOTA approach for imagesynthesis. However, the existing DM cannot perform well on some image-to-imagetranslation (I2I) tasks. Different from image synthesis, some I2I tasks, suchas super-resolution, require generating results in accordance with GT images.Traditional DMs for image synthesis require extensive iterations and largedenoising models to estimate entire images, which gives their strong generativeability but also leads to artifacts and inefficiency for I2I. To tackle thischallenge, we propose a simple, efficient, and powerful DM framework for I2I,called DiffI2I. Specifically, DiffI2I comprises three key components: a compactI2I prior extraction network (CPEN), a dynamic I2I transformer (DI2Iformer),and a denoising network. We train DiffI2I in two stages: pretraining and DMtraining. For pretraining, GT and input images are fed into CPEN$_{S1}$ tocapture a compact I2I prior representation (IPR) guiding DI2Iformer. In thesecond stage, the DM is trained to only use the input images to estimate thesame IRP as CPEN$_{S1}$. Compared to traditional DMs, the compact IPR enablesDiffI2I to obtain more accurate outcomes and employ a lighter denoising networkand fewer iterations. Through extensive experiments on various I2I tasks, wedemonstrate that DiffI2I achieves SOTA performance while significantly reducingcomputational burdens.

DiffI2I: Efficient Diffusion Model for Image-to-Image Translation