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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Current video-based scene graph generation (VidSGG) methods have been foundto perform poorly on predicting predicates that are less represented due to theinherent biased distribution in the training data. In this paper, we take acloser look at the predicates and identify that most visual relations (e.g.sit_above) involve both actional pattern (sit) and spatial pattern (above),while the distribution bias is much less severe at the pattern level. Based onthis insight, we propose a decoupled label learning (DLL) paradigm to addressthe intractable visual relation prediction from the pattern-level perspective.Specifically, DLL decouples the predicate labels and adopts separateclassifiers to learn actional and spatial patterns respectively. The patternsare then combined and mapped back to the predicate. Moreover, we propose aknowledge-level label decoupling method to transfer non-target knowledge fromhead predicates to tail predicates within the same pattern to calibrate thedistribution of tail classes. We validate the effectiveness of DLL on thecommonly used VidSGG benchmark, i.e. VidVRD. Extensive experiments demonstratethat the DLL offers a remarkably simple but highly effective solution to thelong-tailed problem, achieving the state-of-the-art VidSGG performance.

Taking A Closer Look at Visual Relation: Unbiased Video Scene Graph Generation with Decoupled Label Learning