Decoupled Spatial-temporal Predicting Model for Weakly Supervised Action Localization

Wang Gui-qin

doi:10.1016/j.knosys.2025.115241

Decoupled Spatial-temporal Predicting Model for Weakly Supervised Action Localization

2026-01-07

0

PRE

AI

DOI：10.1016/j.knosys.2025.115241

Original

Original request for help

Save

Abstract

En 中文

• Unsupervisedly predict the potential action boundaries for WSAL • Build a novel framework to learn temporal variations in VLM • Build a residual hierarchical spatio-temporal predictive model • Extensive experiments prove the effectiveness of our method

Decoupled Spatial-temporal Predicting Model for Weakly Supervised Action Localization

Abstract

Journal

Organization