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/

Zendy Plus

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

Zendy Tools

Zendy Open

Recent studies have demonstrated the power of recurrent neural networks formachine translation, image captioning and speech recognition. For the task ofcapturing temporal structure in video, however, there still remain numerousopen research questions. Current research suggests using a simple temporalfeature pooling strategy to take into account the temporal aspect of video. Wedemonstrate that this method is not sufficient for gesture recognition, wheretemporal information is more discriminative compared to general videoclassification tasks. We explore deep architectures for gesture recognition invideo and propose a new end-to-end trainable neural network architectureincorporating temporal convolutions and bidirectional recurrence. Our maincontributions are twofold; first, we show that recurrence is crucial for thistask; second, we show that adding temporal convolutions leads to significantimprovements. We evaluate the different approaches on the Montalbano gesturerecognition dataset, where we achieve state-of-the-art results.

Beyond Temporal Pooling: Recurrence and Temporal Convolutions for Gesture Recognition in Video