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

In this paper, we are tackling the weakly-supervised referring expressiongrounding task, for the localization of a referent object in an image accordingto a query sentence, where the mapping between image regions and queries arenot available during the training stage. In traditional methods, an objectregion that best matches the referring expression is picked out, and then thequery sentence is reconstructed from the selected region, where thereconstruction difference serves as the loss for back-propagation. The existingmethods, however, conduct both the matching and the reconstructionapproximately as they ignore the fact that the matching correctness is unknown.To overcome this limitation, a discriminative triad is designed here as thebasis to the solution, through which a query can be converted into one ormultiple discriminative triads in a very scalable way. Based on thediscriminative triad, we further propose the triad-level matching andreconstruction modules which are lightweight yet effective for theweakly-supervised training, making it three times lighter and faster than theprevious state-of-the-art methods. One important merit of our work is itssuperior performance despite the simple and neat design. Specifically, theproposed method achieves a new state-of-the-art accuracy when evaluated onRefCOCO (39.21%), RefCOCO+ (39.18%) and RefCOCOg (43.24%) datasets, that is4.17%, 4.08% and 7.8% higher than the previous one, respectively.

Discriminative Triad Matching and Reconstruction for Weakly Referring Expression Grounding