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

3D Convolutional Neural Network (3D CNN) captures spatial and temporalinformation on 3D data such as video sequences. However, due to the convolutionand pooling mechanism, the information loss seems unavoidable. To improve thevisual explanations and classification in 3D CNN, we propose two approaches; i)aggregate layer-wise global to local (global-local) discrete gradients usingtrained 3DResNext network, and ii) implement attention gating network toimprove the accuracy of the action recognition. The proposed approach intendsto show the usefulness of every layer termed as global-local attention in 3DCNN via visual attribution, weakly-supervised action localization, and actionrecognition. Firstly, the 3DResNext is trained and applied for actionclassification using backpropagation concerning the maximum predicted class.The gradients and activations of every layer are then up-sampled. Later,aggregation is used to produce more nuanced attention, which points out themost critical part of the predicted class's input videos. We use contourthresholding of final attention for final localization. We evaluate spatial andtemporal action localization in trimmed videos using fine-grained visualexplanation via 3DCam. Experimental results show that the proposed approachproduces informative visual explanations and discriminative attention.Furthermore, the action recognition via attention gating on each layer producesbetter classification results than the baseline model.

Weakly-Supervised Action Localization and Action Recognition using Global-Local Attention of 3D CNN