VideoLSTM convolves, attends and flows for action recognition

• Published in: Computer vision and image understanding Vol. 166; pp. 41 - 50
• Main Authors: Li, Zhenyang, Gavrilyuk, Kirill, Gavves, Efstratios, Jain, Mihir, Snoek, Cees G.M.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.01.2018
• Subjects: Action recognition; Attention; LSTM; Video representation; Action recognition; LSTM; Attention; Video representation

•To exploit both the spatial and temporal correlations in a video, we hardwire convolutions in the soft-Attention LSTM architecture.•We introduce motion-based attention which guides better the attention towards the relevant spatial-temporal locations of the actions.•We demonstrate how the attention...