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

Structure-guided image completion aims to inpaint a local region of an imageaccording to an input guidance map from users. While such a task enables manypractical applications for interactive editing, existing methods often struggleto hallucinate realistic object instances in complex natural scenes. Such alimitation is partially due to the lack of semantic-level constraints insidethe hole region as well as the lack of a mechanism to enforce realistic objectgeneration. In this work, we propose a learning paradigm that consists ofsemantic discriminators and object-level discriminators for improving thegeneration of complex semantics and objects. Specifically, the semanticdiscriminators leverage pretrained visual features to improve the realism ofthe generated visual concepts. Moreover, the object-level discriminators takealigned instances as inputs to enforce the realism of individual objects. Ourproposed scheme significantly improves the generation quality and achievesstate-of-the-art results on various tasks, including segmentation-guidedcompletion, edge-guided manipulation and panoptically-guided manipulation onPlaces2 datasets. Furthermore, our trained model is flexible and can supportmultiple editing use cases, such as object insertion, replacement, removal andstandard inpainting. In particular, our trained model combined with a novelautomatic image completion pipeline achieves state-of-the-art results on thestandard inpainting task.

Structure-Guided Image Completion with Image-level and Object-level Semantic Discriminators