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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Zero-shot skeleton-based action recognition aims to recognize actions ofunseen categories after training on data of seen categories. The key is tobuild the connection between visual and semantic space from seen to unseenclasses. Previous studies have primarily focused on encoding sequences into asingular feature vector, with subsequent mapping the features to an identicalanchor point within the embedded space. Their performance is hindered by 1) theignorance of the global visual/semantic distribution alignment, which resultsin a limitation to capture the true interdependence between the two spaces. 2)the negligence of temporal information since the frame-wise features with richaction clues are directly pooled into a single feature vector. We propose a newzero-shot skeleton-based action recognition method via mutual information (MI)estimation and maximization. Specifically, 1) we maximize the MI between visualand semantic space for distribution alignment; 2) we leverage the temporalinformation for estimating the MI by encouraging MI to increase as more framesare observed. Extensive experiments on three large-scale skeleton actiondatasets confirm the effectiveness of our method. Code:https://github.com/YujieOuO/SMIE.

Zero-shot Skeleton-based Action Recognition via Mutual Information Estimation and Maximization