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

Diffusion models (DMs) have gained prominence due to their ability togenerate high-quality, varied images, with recent advancements in text-to-imagegeneration. The research focus is now shifting towards the controllability ofDMs. A significant challenge within this domain is localized editing, wherespecific areas of an image are modified without affecting the rest of thecontent. This paper introduces LIME for localized image editing in diffusionmodels that do not require user-specified regions of interest (RoI) oradditional text input. Our method employs features from pre-trained methods anda simple clustering technique to obtain precise semantic segmentation maps.Then, by leveraging cross-attention maps, it refines these segments forlocalized edits. Finally, we propose a novel cross-attention regularizationtechnique that penalizes unrelated cross-attention scores in the RoI during thedenoising steps, ensuring localized edits. Our approach, without re-trainingand fine-tuning, consistently improves the performance of existing methods invarious editing benchmarks.

LIME: Localized Image Editing via Attention Regularization in Diffusion Models