时序增强的视频动作识别方法  

Video-Based Temporal Enhanced Action Recognition

作　　者：张浩博 付冬梅[1,3] 周珂[4] ZHANG Haobo;FU Dongmei;ZHOU Ke(School of Automation and Electrical Engineering,University of Science and Technology Beijing,Beijing 100083;Shunde Graduate School,University of Science and Technology Beijing,Foshan 528399;Beijing Engineering Research Center of Industrial Spectrum Imaging,University of Science and Technology Beijing,Beijing 100083;School of Advanced Engineering,University of Science and Technology Beijing,Beijing 100083)

机构地区：[1]北京科技大学自动化学院,北京100083 [2]北京科技大学顺德研究生院,佛山528399 [3]北京科技大学北京市工业波谱成像工程中心,北京100083 [4]北京科技大学高等工程师学院,北京100083

出　　处：《模式识别与人工智能》2020年第10期951-958,共8页Pattern Recognition and Artificial Intelligence

摘　　要：针对视频动作识别中的时空建模问题,在深度学习框架下提出基于融合时空特征的时序增强动作识别方法.首先对输入视频应用稀疏时序采样策略,适应视频时长变化,降低视频级别时序建模成本.在识别阶段计算相邻特征图间的时序差异,以差异计算结果增强特征级别的运动信息.最后,利用残差结构与时序增强结构的组合方式提升网络整体时空建模能力.实验表明,文中算法在UCF101、HMDB51数据集上取得较高准确率,并在实际工业操作动作识别场景下,以较小的网络规模达到较优的识别效果.Aiming at the spatio-temporal modeling in video action recognition,a temporal enhanced action recognition algorithm based on fused spatio-temporal features is proposed under the deep learning framework.To lower the cost of video-level temporal modeling,a sparse sampling strategy is employed to adapt to video duration changes.In the recognition stage,temporal difference between adjacent feature maps is calculated to enhance the motion information in the feature level.The combination of residual structure and temporal enhanced structure is introduced to further improve the representation ability of the network.Experimental results show that the proposed algorithm obtains higher accuracy on UCF101 and HMDB51 datasets and achieves better results in the actual industrial operation recognition scene with a smaller network scale.

关 键 词：动作识别 深度学习 时序增强结构 工业监控视频 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]