基于深度图的人体动作分类自适应算法Adaptive Algorithm for Human Motion Classification Based on Depth Map
蒋韦晔,刘成明
摘要(Abstract)：
由于深度相机成本的降低,越来越多的研究人员使用RGB-D(red, green, blue and depth)视频进行人类动作识别(human activity recognition,HAR)。使用深度运动图的局部二值模式进行特征提取,利用自适应差分进化极限学习机(self-adaptive differential evolution extreme learning machine,SaDE-ELM)用于动作分类,其中隐藏节点的学习参数通过自适应差分进化的方法进行修改。为了验证所提出方法的有效性,用3个公共数据集(MSR Action3D,MSRDaily Activity3D,MSRGesture3D)进行了实验。仿真结果表明,该方法优于基于内核的极限学习机(kernel extreme learning machine,KELM)的方法。
关键词(KeyWords)： 人类动作识别;深度运动图;差分进化;自适应差分进化极限学习机
基金项目(Foundation): 国家自然科学青年基金项目(6140240)
作者(Author): 蒋韦晔,刘成明
DOI: 10.13705/j.issn.1671-6841.2019584
参考文献(References)：
- [1] YANG X,ZHANG C,TIAN Y.Recognizing actions using depth motion maps-based histograms of oriented gradients[C]//Proceedings of the 20th ACM International Conference on Multimedia.New York,2012:1057-1060.
- [2] 程万里.基于深度数据特征融合的人体动作识别[D].郑州:郑州大学,2018.CHENG W L.Human action recognition based on depth data feature fusion[D].Zhengzhou:Zhengzhou University,2018.
- [3] WONG C M,VONG C M,WONG P K,et al.Kernel-based multilayer extreme learning machines for representation learning[J].IEEE transactions on neural networks and learning systems,2018,29(3):757-762.
- [4] 王杰,刘向晴.彩色图像分割的FCM预分类核极限学习机方法[J].郑州大学学报(理学版),2018,50(2):75-80.WANG J,LIU X Q.FCM pre-classification kernel extreme learning machine algorithm of color image segmentation[J].Journal of Zhengzhou university(natural science edition),2018,50(2):75-80.
- [5] ZHU Q Y,QIN A K,SUGANTHAN P N,et al.Evolutionary extreme learning machine[J].Pattern recognition,2005,38(10):1759-1763.
- [6] MSRaction3D dataset.(2018-01-05)[2019-02-10].http://research.microsoft.com/enus/um/people/zliu/ actionrecorsrc/.
- [7] MSRdailyActivity3D dataset.(2018-09-07)[2019-02-10].http://users.eecs.northwestern.edu/～jwa368/ my_data.html.
- [8] DING W W,LIU K,FU X J,et al.Profile HMMs for skeleton-based human action recognition[J].Signal processing:image communication,2016,42:109-119.
- [9] KARAMIZADEH S,ABDULLAH S M,MANAF A A,et al.An overview of principal component analysis[J].Journal of signal and information processing,2013,4(3):173-175.
- [10] CHEN C,JAFARI R.Action recognition from depth sequences using depth motion maps-based local binary patterns[C]//2015 15th IEEE Winter Conference on Applications of Computer Vision.Waikoloa,2015:1092-1099.
- [11] NUNES U M,FARIA D R,PEIXOTO P.A human activity recognition framework using max-min features and key poses with differential evolution random forests classifier[J].Pattern recognition letters,2017,99:21-31.