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

For the two-stream style methods in action recognition, fusing the twostreams' predictions is always by the weighted averaging scheme. This fusionmethod with fixed weights lacks of pertinence to different action videos andalways needs trial and error on the validation set. In order to enhance theadaptability of two-stream ConvNets and improve its performance, an end-to-endtrainable gated fusion method, namely gating ConvNet, for the two-streamConvNets is proposed in this paper based on the MoE (Mixture of Experts)theory. The gating ConvNet takes the combination of feature maps from the samelayer of the spatial and the temporal nets as input and adopts ReLU (RectifiedLinear Unit) as the gating output activation function. To reduce theover-fitting of gating ConvNet caused by the redundancy of parameters, a newmulti-task learning method is designed, which jointly learns the gating fusionweights for the two streams and learns the gating ConvNet for actionclassification. With our gated fusion method and multi-task learning approach,a high accuracy of 94.5% is achieved on the dataset UCF101.

Learning Gating ConvNet for Two-Stream based Methods in Action Recognition