Mean Polynomial Kernel and Its Application to Vector Sequence Recognition
Mean Polynomial Kernel and Its Application to Vector Sequence Recognition
复制标题
DOI：
10.1587/transinf.e97.d.1855
复制
发表时间：
2014-07
期刊：
影响因子：
--
通讯作者：
Raissa Relator;Yoshihiro Hirohashi;Eisuke Ito;Tsuyoshi Kato
中科院分区：
文献类型：
--
作者：
Raissa Relator;Yoshihiro Hirohashi;Eisuke Ito;Tsuyoshi Kato
计算机视觉和脑机接口研究中的分类任务已经提出了一些应用，如生物识别和认知训练。然而，确定合适的数据表示一直具有挑战性，最近的方法已经偏离了每个数据样本的一个向量的熟悉形式。本文考虑了向量集之间的内核，平均多项式内核，最近的研究，数据近似的线性子空间，特别是格拉斯曼流形上的方法。内核支持向量序列作为输入。我们讨论了如何内核可以与格拉斯曼投影内核，并提供实验结果显示它如何优于现有的基于子空间的方法格拉斯曼流形。
Classification tasks in computer vision and brain-computer interface research have presented several applications such as biometrics and cognitive training. However, determining suitable data representation has been challenging, and recent approaches have deviated from the familiar form of one vector for each data sample. This paper considers a kernel between vector sets, the mean polynomial kernel, motivated by recent studies where data are approximated by linear subspaces, in particular, methods on Grassmann manifolds. The kernel supports vector sequences as input. We discuss how the kernel can be associated with the Grassmann Projection kernel, and provide experimental results showing how it outperforms existing subspace-based methods on Grassmann manifolds.