Enhanced Grassmann discriminant analysis with randomized time warping for motion recognition
Enhanced Grassmann discriminant analysis with randomized time warping for motion recognition
复制标题
DOI：
10.1016/j.patcog.2019.107028
复制
发表时间：
2020-01
期刊：
影响因子：
--
通讯作者：
L. S. Souza;B. Gatto;Jing-Hao Xue;K. Fukui
中科院分区：
文献类型：
--
作者：
L. S. Souza;B. Gatto;Jing-Hao Xue;K. Fukui
通过对Grassmann判别分析(GDA)框架的扩展，提出了一种运动序列分类框架。GDA的一个问题是它的判别空间不一定是最优的。当利用随机时间扭曲(RTW)的子空间表示时，这一限制变得更加突出。RTW是一种序列表示法，可以有效地用低维子空间来建模运动的时间信息，将两个序列的比较问题简化为两个子空间的比较问题。该算法的核心思想是先将类子空间投影到广义差子空间上，然后再映射到Grassmann流形上。GDS投影可以去掉向量空间中子空间的重叠分量，使它们近乎正交化。因此，与原始集合相比，正交化类子空间的字典在Grassmann流形中产生一组更具判别性的数据点。这组数据点可以进一步增强GDA的判别能力。通过使用公开可用的剑桥手势、KTH动作和UCF体育数据集进行的运动识别实验，我们证明了所提出的RTW+eGDA框架的有效性。
This paper proposes a framework for classifying motion sequences, by extending the framework of Grassmann discriminant analysis (GDA). A problem of GDA is that its discriminant space is not necessarily optimal. This limitation becomes even more prominent when utilizing the subspace representation of randomized time warping (RTW). RTW is a sequence representation that can effectively model a motion’s temporal information by a low-dimensional subspace, simplifying the problem of comparing two sequences to that of comparing two subspaces. The key idea of the proposed enhanced GDA is projecting class subspaces onto a generalized difference subspace before mapping them on a Grassmann manifold. The GDS projection can remove overlapping components of the subspaces in the vector space, nearly orthogonalizing them. Consequently, a dictionary of orthogonalized class subspaces produces a set of more discriminant data points in the Grassmann manifold, in comparison with the original set. This set of data points can further enhance the discriminant ability of GDA. We demonstrate the validity of the proposed framework, RTW+eGDA, through experiments on motion recognition using the publicly available Cambridge gesture, KTH action, and UCF sports datasets.