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 proposal-free referring expressiongrounding task, aiming at localizing the target object according to a querysentence, without relying on off-the-shelf object proposals. Existingproposal-free methods employ a query-image matching branch to select thehighest-score point in the image feature map as the target box center, with itswidth and height predicted by another branch. Such methods, however, fail toutilize the contextual relation between the target and reference objects, andlack interpretability on its reasoning procedure. To solve these problems, wepropose an iterative shrinking mechanism to localize the target, where theshrinking direction is decided by a reinforcement learning agent, with allcontents within the current image patch comprehensively considered. Beside, thesequential shrinking process enables to demonstrate the reasoning about how toiteratively find the target. Experiments show that the proposed method booststhe accuracy by 4.32% against the previous state-of-the-art (SOTA) method onthe RefCOCOg dataset, where query sentences are long and complex, with manytargets referred by other reference objects.

Iterative Shrinking for Referring Expression Grounding Using Deep Reinforcement Learning