Z
Zhengming Ma
Researcher at Sun Yat-sen University
Publications - 43
Citations - 270
Zhengming Ma is an academic researcher from Sun Yat-sen University. The author has contributed to research in topics: Dimensionality reduction & Subspace topology. The author has an hindex of 7, co-authored 43 publications receiving 220 citations.
Papers
More filters
Journal ArticleDOI
Face recognition based on the fusion of global and local HOG features of face images
TL;DR: A novel framework for face recognition based on the fusion of global and local HOG features has been proposed and shows that, in comparison with 12 state-of-the-art approaches of face recognition, the proposed method achieves the highest average recognition rate.
Journal ArticleDOI
Local Coordinates Alignment With Global Preservation for Dimensionality Reduction
Jing Chen,Zhengming Ma,Yang Liu +2 more
TL;DR: The proposed alignment method, called landmarks-based alignment, can produce a closed-form solution without any constraints, while most previous alignment-based methods impose the unit covariance constraint, which will result in the deficiency of global metrics and undesired rescaling of the manifold.
Journal ArticleDOI
Grassmann manifold for nearest points image set classification
TL;DR: A novel approach is proposed by exploiting the Projection kernel that explicitly maps the subspaces from the Grassmann manifold to a Reproducing Kernel Hilbert Space (RKHS) where the Euclidean geometry applies, and then, by modeling the points on RKHS as affine hulls, the Euclidan distance between the nearest points of two hulls can be used for classification.
Journal ArticleDOI
Locally linear embedding: a review
Jing Chen,Jing Chen,Zhengming Ma +2 more
TL;DR: By generalizing different tactics in various extensions related to different stages of LLE and evaluating their performances, several promising directions for future research have been suggested.
Journal ArticleDOI
On the Equivalence of HLLE and LTSA
TL;DR: This paper proves that, although HLLE and LTSA uses different methods to solve the alignment equation, their solutions are exactly the same, provided that HLLE adopts the same method as LTSA to construct the neighborhoods.