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

Image Retrieval is a fundamental task of obtaining images similar to thequery one from a database. A common image retrieval practice is to firstlyretrieve candidate images via similarity search using global image features andthen re-rank the candidates by leveraging their local features. Previouslearning-based studies mainly focus on either global or local imagerepresentation learning to tackle the retrieval task. In this paper, we abandonthe two-stage paradigm and seek to design an effective single-stage solution byintegrating local and global information inside images into compact imagerepresentations. Specifically, we propose a Deep Orthogonal Local and Global(DOLG) information fusion framework for end-to-end image retrieval. Itattentively extracts representative local information with multi-atrousconvolutions and self-attention at first. Components orthogonal to the globalimage representation are then extracted from the local information. At last,the orthogonal components are concatenated with the global representation as acomplementary, and then aggregation is performed to generate the finalrepresentation. The whole framework is end-to-end differentiable and can betrained with image-level labels. Extensive experimental results validate theeffectiveness of our solution and show that our model achieves state-of-the-artimage retrieval performances on Revisited Oxford and Paris datasets.

DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Features