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

Video highlight detection is a crucial yet challenging problem that aims toidentify the interesting moments in untrimmed videos. The key to this task liesin effective video representations that jointly pursue two goals,\textit{i.e.}, cross-modal representation learning and fine-grained featurediscrimination. In this paper, these two challenges are tackled by not onlyenriching intra-modality and cross-modality relations for representationmodeling but also shaping the features in a discriminative manner. Our proposedmethod mainly leverages the intra-modality encoding and cross-modalityco-occurrence encoding for fully representation modeling. Specifically,intra-modality encoding augments the modality-wise features and dampensirrelevant modality via within-modality relation learning in both audio andvisual signals. Meanwhile, cross-modality co-occurrence encoding focuses on theco-occurrence inter-modality relations and selectively captures effectiveinformation among multi-modality. The multi-modal representation is furtherenhanced by the global information abstracted from the local context. Inaddition, we enlarge the discriminative power of feature embedding with ahard-pairs guided contrastive learning (HPCL) scheme. A hard-pairs samplingstrategy is further employed to mine the hard samples for improving featurediscrimination in HPCL. Extensive experiments conducted on two benchmarksdemonstrate the effectiveness and superiority of our proposed methods comparedto other state-of-the-art methods.

Probing Visual-Audio Representation for Video Highlight Detection via Hard-Pairs Guided Contrastive Learning