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

Weakly-supervised Temporal Action Localization (WS-TAL) methods learn tolocalize temporal starts and ends of action instances in a video under onlyvideo-level supervision. Existing WS-TAL methods rely on deep features learnedfor action recognition. However, due to the mismatch between classification andlocalization, these features cannot distinguish the frequently co-occurringcontextual background, i.e., the context, and the actual action instances. Weterm this challenge action-context confusion, and it will adversely affect theaction localization accuracy. To address this challenge, we introduce aframework that learns two feature subspaces respectively for actions and theircontext. By explicitly accounting for action visual elements, the actioninstances can be localized more precisely without the distraction from thecontext. To facilitate the learning of these two feature subspaces with onlyvideo-level categorical labels, we leverage the predictions from both spatialand temporal streams for snippets grouping. In addition, an unsupervisedlearning task is introduced to make the proposed module focus on miningtemporal information. The proposed approach outperforms state-of-the-art WS-TALmethods on three benchmarks, i.e., THUMOS14, ActivityNet v1.2 and v1.3datasets.

Weakly Supervised Temporal Action Localization Through Learning Explicit Subspaces for Action and Context