全文总字数：2032字

1. 毕业设计（论文）的内容、要求、设计方案、规划等

本研究目标是利用多通道卷积神经网络并融合输出多任务标签，为视频分类和视频描述奠定基础。

论文要求按照标准格式书写，并主要包括以下内容： 前言：介绍国内外此方面工作的研究进展，并叙述该系统的研究意义。

此章是毕业论文的核心部分。

2. 参考文献（不低于12篇）

two-stream convolutional networks for action recognition in videos (nips, 2014)convolutional two-stream network fusion for video action recognition (cvpr, 2016)multi-stream multi-class fusion of deep networks for video classification (acm multimedia, 2016)a flexible cnn framework for multi-label image classification (tpami, 2016)multi-region two-stream r-cnn for action detection (eccv, 2016)[1] m. baccouche, f. mamalet, c. wolf, c. garcia, anda. baskurt. action classification in soccer videos withlong short-term memory recurrent neural networks. in proc.icann, pages 154159, thessaloniki, greece, 2010. 2

[2] m. baccouche, f. mamalet, c. wolf, c. garcia, anda. baskurt. sequential deep learning for human actionrecognition. in 2nd international workshop on human behavior understanding (hbu), pages 2939, nov. 2011. 1,2

[3] y. bengio, p. simard, and p. frasconi. learning long-termdependencies with gradient descent is difficult. ieee trans.on neural networks, 5(2):157166, 1994. 2