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

Diverse video captioning aims to generate a set of sentences to describe thegiven video in various aspects. Mainstream methods are trained with independentpairs of a video and a caption from its ground-truth set without exploiting theintra-set relationship, resulting in low diversity of generated captions.Different from them, we formulate diverse captioning into asemantic-concept-guided set prediction (SCG-SP) problem by fitting thepredicted caption set to the ground-truth set, where the set-level relationshipis fully captured. Specifically, our set prediction consists of two synergistictasks, i.e., caption generation and an auxiliary task of concept combinationprediction providing extra semantic supervision. Each caption in the set isattached to a concept combination indicating the primary semantic content ofthe caption and facilitating element alignment in set prediction. Furthermore,we apply a diversity regularization term on concepts to encourage the model togenerate semantically diverse captions with various concept combinations. Thesetwo tasks share multiple semantics-specific encodings as input, which areobtained by iterative interaction between visual features and conceptualqueries. The correspondence between the generated captions and specific conceptcombinations further guarantees the interpretability of our model. Extensiveexperiments on benchmark datasets show that the proposed SCG-SP achievesstate-of-the-art (SOTA) performance under both relevance and diversity metrics.

Set Prediction Guided by Semantic Concepts for Diverse Video Captioning