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

The recognition of behaviors in videos usually requires a combinatorialanalysis of the spatial information about objects and their dynamic actioninformation in the temporal dimension. Specifically, behavior recognition mayeven rely more on the modeling of temporal information containing short-rangeand long-range motions; this contrasts with computer vision tasks involvingimages that focus on the understanding of spatial information. However, currentsolutions fail to jointly and comprehensively analyze short-range motionbetween adjacent frames and long-range temporal aggregations at large scales invideos. In this paper, we propose a novel behavior recognition method based onthe integration of multigranular (IMG) motion features. In particular, weachieve reliable motion information modeling through the synergy of a channelattention-based short-term motion feature enhancement module (CMEM) and acascaded long-term motion feature integration module (CLIM). We evaluate ourmodel on several action recognition benchmarks such as HMDB51,Something-Something and UCF101. The experimental results demonstrate that ourapproach outperforms the previous state-of-the-art methods, which confirms itseffectiveness and efficiency.

Behavior Recognition Based on the Integration of Multigranular Motion Features