13 基于低秩稀疏优化的交互行为识别

马欣;李建军;

• 1:内蒙古科技大学信息工程学院

• 
  

摘要(Abstract)：

针对复杂的交互行为在识别过程中特征维数高、信息冗余大导致的识别率较低的问题,文章提出了基于深度图像和RGB图像特征的低秩稀疏融合算法.采用基于稀疏表示的空间金字塔模型对静态特征和动态特征进行编码融合.最后采用低秩稀疏优化方法进行行为识别,该方法在公共数据集CAD-60和MSR Action Pairs上进行实验验证,取得了理想的效果.

关键词(KeyWords)： 交互行为识别;特征融合;稀疏;低秩

基金项目(Foundation): 国家自然科学基金资助项目(62066036);; 内蒙古自治区自然科学基金资助项目(2018MS06018);; 内蒙古自治区杰青培育基金资助项目(2018JQ02)

作者(Author): 马欣;李建军;

Email:

DOI: 10.16559/j.cnki.2095-2295.2021.04.013

参考文献(References)：

• [1] Zhang J,Xu C,Jing P,et al. A tensor-driven temporal correlation model for video sequence classification[J].IEEE Signal Process Lett,2016,23(9):1246. [2]陈昌红，张杰，刘峰．双人交互行为的稀疏表征方法[J]．模式识别与人工智能，2016,29(5):464. [3] Zheng C,Hou Y,Zhang J. Improved sparse representation with low-rank representation for robust face recognition[J]. Neurocomputing,2016,198(5):114. [4]陈斌，东一舟，朱晋宁．非控场景下主成分稀疏表示与低秩分解的人脸识别[J]．液晶与显示，2019,34(8):816. [5]叶茂权．基于视频的人体行为识别的理论与方法研究[D]．成都:电子科技大学，2018. [6] Candès E J,Li X D,Ma Y,et al. Robust principal component analysis[J]. Journal of the ACM,2011,58(3):#11.[DOL:10. 1145/1970392. 1970395]. [7]孙彬，孔德慧，张雯晖，等．基于深度图像的人体行为识别综述[J]．北京工业大学学报，2018,44(10):1353. [8]范晞，费胜巍，储有兵．基于Canny算子的改进型图像边缘提取算法[J]．自动化与仪表，2019,34(1):41. [9] Canny J. A Computation approach to edge detection[C]//IEEE transactions on pattern analysis and machine intelligence. Washington DC,USA:IEEE Computer Society,1986:679. [10]张磊．基于改进LBP纹理图像特征提取与分类研究[D]．哈尔滨:哈尔滨工程大学，2019. [11] Liu L,Chen J,Fieguth P,et al. From Bo W to CNN:Two decades of texture representation for texture classification[J]. International Journal of Computer Vision,2019,127(1):74. [12] Ojala T,Pietikainen M,Maenpaa T. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[C]//IEEE transactions on pattern analysis and machine intelligence. Washington DC,USA:IEEE Computer Society,2002:987. [13]徐海燕．复杂环境下行为识别特征提取方法研究[D]．南京:东南大学，2016. [14] Yang J,Yu K,Gong Y,et al. Linear spatial pyramid matching using sparse coding for image classification[C]//Computer society conference on computer vision and pattern recognition. Florida,USA:IEEE,2009:1794. [15]向程谕．RGB-D图像的特征提取与分类方法研究[D]．湘潭:湘潭大学，2017. [16]木立生．基于特征融合的低秩恢复稀疏表示人脸识别[D]．太原:太原理工大学，2017. [17]李欢．基于低秩稀疏理论的图像去噪算法研究[D]．南京:南京邮电大学，2018. [18] Chen M,Lin Z,Ma Y,et al. The augmented lagrange multiplier method for exact recovery of corrupted lowrank matrices[J]. Eprint Arxiv,2010(9):58. [19] Gaglio S,Re G L,Morana M. Human activity recognition process using 3-D posture data[C]//IEEE transaction on pattern analysis and machine intelligence. Washington DC,USA:IEEE computer society,2014:2168. [20] Koppula H S,Gupta R,Saxena A. Learning human activities and object affordances from RGB-D videos[J].The International Journal of Robotics Research,2013,32(8):951. [21]王永雄，曾艳，李璇，等．融合交互信息和能量特征的三维复杂人体行为识别[J]．小型微型计算机系统，2018,39(8):1828. [22] Chen C,Jafari R,Kehtarnavaz N. Action recognition from depth sequences using depth motion maps-based local binary patterns[J]. Applications of Computer Vision,2015,12(7):251. [23] Rahmani H,Mahmood A,Du Q H,et al. Real time action recognition using histograms of depth gradients and random decision forests[C]//IEEE transaction on pattern analysis and machine intelligence. Washington DC,USA:IEEE Computer Society,2014:626. [24] Wang J,Liu Z,Wu Y,et al. Mining actionlet ensemble for action recognition with depth cameras[C]//IEEE transaction on pattern analysis and machine intelligence.Washington D C,USA:IEEE Computer Society,2012:1290. [25] Oreifej O,Liu Z. HON4D:histogram of ori-ented 4D normal for activity recognition from depth sequence[C]//Conference on computer vision and pattern recognition. New York,USA:IEEE,2013:723. [26] Wang J,Wu Y. Learning Maximum Margin temporal warping for action recognition[C]//IEEE transaction on pattern analysis and machine intelligence. Washington DC,USA:IEEE Computer Society,2013:2688. [27] Seddik B,Gazzah S,Najoua E. Human-action recognition using a multi-layered fusion scheme of kinect modalities[J]. IET Computer Vision,2017,11(7):530. [28] Li Z,Yap K H. An efficient approach for scene categorization based on discriminative codebook learning in bag-of-words framework[J]. Image and Vision Computing,2013,31(10):748.