Depth And Skeleton Based View-invariant Human Action Recognition | IEEE Conference Publication | IEEE Xplore