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

Person clustering with multi-modal clues, including faces, bodies, andvoices, is critical for various tasks, such as movie parsing and identity-basedmovie editing. Related methods such as multi-view clustering mainly projectmulti-modal features into a joint feature space. However, multi-modal cluefeatures are usually rather weakly correlated due to the semantic gap from themodality-specific uniqueness. As a result, these methods are not suitable forperson clustering. In this paper, we propose a Relation-Aware Distributionrepresentation Network (RAD-Net) to generate a distribution representation formulti-modal clues. The distribution representation of a clue is a vectorconsisting of the relation between this clue and all other clues from allmodalities, thus being modality agnostic and good for person clustering.Accordingly, we introduce a graph-based method to construct distributionrepresentation and employ a cyclic update policy to refine distributionrepresentation progressively. Our method achieves substantial improvements of+6% and +8.2% in F-score on the Video Person-Clustering Dataset (VPCD) andVoxCeleb2 multi-view clustering dataset, respectively. Codes will be releasedpublicly upon acceptance.

Relation-Aware Distribution Representation Network for Person Clustering with Multiple Modalities