Heterogeneous massive feature fusion on grassmannian manifold

2nd Annual International Conference on Information System and Artificial Intelligence

物理学;计算机科学

Huang, Haichao^1 ; Liu, Hongning^1 ; Kong, Xiaoyun^2 ; Lou, Xingdan^3 ; Wang, Zepeng^4

State Grid Zhejiang Information and Telecommunication Company, China^1

State Grid Zhejiang Electric Power Company, China^2

Zhejiang Huayun Information Technology CO. LTD, China^3

Department of CSIE, Hefei University of Technology, China^4

关键词: Curse of dimensionality;    Grassmann manifold;    Grassmannian manifolds;    Hausdorff distance;    Multi-modal dataset;    Multimodal features;    Selection criterion;    State of the art;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/887/1/012066/pdf
DOI  :  10.1088/1742-6596/887/1/012066

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Two issues remain unsolved on utilizing multimodal features for pattern recognition: the missing features and the curse of dimensionality. In this paper, we address the two issues by fusing the multimodal features on the Grassmann manifold. By defining grouping constrains on multimodal features, each multimodal feature vector is grouped into a set of subspaces, and is further represented as a point on the Grassmann manifold. To deal with missing features, L2-Hausdorff distance, a metric to compare multimodal feature vectors with different number of subspaces, is computed, and a kernel matrix can be obtained accordingly. Based on the kernel matrix, two feature selection criterions, one supervised and one unsupervised, are proposed to obtain a few representative features in the kernel space. Thus, the curse of dimensionality is alleviated. Experimental results on three multimodal dataset show the proposed feature fusion can outperforms the state-of -the-art by higher accuracy.