%0 Conference Proceedings
%T Attention-Based Temporal Weighted Convolutional Neural Network for Action Recognition
%+ Xi'an Jiaotong University (Xjtu)
%+ HERE technologies
%+ Microsoft Research
%A Zang, Jinliang
%A Wang, Le
%A Liu, Ziyi
%A Zhang, Qilin
%A Hua, Gang
%A Zheng, Nanning
%Z Part 2: Deep Learning
%< avec comité de lecture
%( IFIP Advances in Information and Communication Technology
%B 14th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI)
%C Rhodes, Greece
%Y Lazaros Iliadis
%Y Ilias Maglogiannis
%Y Vassilis Plagianakos
%I Springer International Publishing
%3 Artificial Intelligence Applications and Innovations
%V AICT-519
%P 97-108
%8 2018-05-25
%D 2018
%R 10.1007/978-3-319-92007-8_9
%K Action recognition
%K Attention model
%K Convolutional neural networks
%K Video-level prediction
%K Temporal weighting
%Z Computer Science [cs]Conference papers
%X Research in human action recognition has accelerated significantly since the introduction of powerful machine learning tools such as Convolutional Neural Networks (CNNs). However, effective and efficient methods for incorporation of temporal information into CNNs are still being actively explored in the recent literature. Motivated by the popular recurrent attention models in the research area of natural language processing, we propose the Attention-based Temporal Weighted CNN (ATW), which embeds a visual attention model into a temporal weighted multi-stream CNN. This attention model is simply implemented as temporal weighting yet it effectively boosts the recognition performance of video representations. Besides, each stream in the proposed ATW frame- work is capable of end-to-end training, with both network parameters and temporal weights optimized by stochastic gradient descent (SGD) with backpropagation. Our experiments show that the proposed attention mechanism contributes substantially to the performance gains with the more discriminative snippets by focusing on more relevant video segments.
%G English
%Z TC 12
%Z WG 12.5
%2 https://inria.hal.science/hal-01821048/document
%2 https://inria.hal.science/hal-01821048/file/467708_1_En_9_Chapter.pdf
%L hal-01821048
%U https://inria.hal.science/hal-01821048
%~ IFIP
%~ IFIP-AICT
%~ IFIP-TC
%~ IFIP-WG
%~ IFIP-TC12
%~ IFIP-AIAI
%~ IFIP-WG12-5
%~ IFIP-AICT-519