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

Compressed video action recognition has recently drawn growing attention,since it remarkably reduces the storage and computational cost via replacingraw videos by sparsely sampled RGB frames and compressed motion cues (e.g.,motion vectors and residuals). However, this task severely suffers from thecoarse and noisy dynamics and the insufficient fusion of the heterogeneous RGBand motion modalities. To address the two issues above, this paper proposes anovel framework, namely Attentive Cross-modal Interaction Network with MotionEnhancement (MEACI-Net). It follows the two-stream architecture, i.e. one forthe RGB modality and the other for the motion modality. Particularly, themotion stream employs a multi-scale block embedded with a denoising module toenhance representation learning. The interaction between the two streams isthen strengthened by introducing the Selective Motion Complement (SMC) andCross-Modality Augment (CMA) modules, where SMC complements the RGB modalitywith spatio-temporally attentive local motion features and CMA further combinesthe two modalities with selective feature augmentation. Extensive experimentson the UCF-101, HMDB-51 and Kinetics-400 benchmarks demonstrate theeffectiveness and efficiency of MEACI-Net.

Representation Learning for Compressed Video Action Recognition via Attentive Cross-modal Interaction with Motion Enhancement