Human action recognition using double discriminative sparsity preserving projections and discriminant ridge-based classifier based on the GDWL-l1 graph

作者：

Highlights：

• GDWL- L1 graph captures sparsity, locality and discriminative information of data.

• Two kernels are defined based on the geodesic distance and structural information.

• DDSPP maps high-dimensional data to a sparse, discriminative low-dimensional space.

• DRC classifies data based on the ridge regression and a classification criterion.

摘要

•GDWL- L1 graph captures sparsity, locality and discriminative information of data.•Two kernels are defined based on the geodesic distance and structural information.•DDSPP maps high-dimensional data to a sparse, discriminative low-dimensional space.•DRC classifies data based on the ridge regression and a classification criterion.

论文关键词：Double discriminative sparsity preserving projections,Geodesic distance based weighted LASSO l1-graph,Discriminant ridge-based classifier,Least absolute shrinkage and selection operator (LASSO)

论文评审过程：Received 17 October 2018, Revised 24 June 2019, Accepted 4 September 2019, Available online 6 September 2019, Version of Record 13 September 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.112927