Kernel discriminant analysis for color face recognition
MetadataShow full item record
A lot of algorithms for face recognition have been proposed in current literature and they are mainly for gray images. They will have some limitations when applying to color face images. In this paper, a novel kernel method is developed with an aim to improve the performance for color face recognition. Technically, this method will combine kernel method with Linear Discriminant Analysis for color face models. Intuitively the color space can provide extra discriminant information and the kernel method will provide the nonlinear discriminant capability. Their combination will take both advantages and can improve the recognition performance significantly as demonstrated in the extensive experiments. Experiments are done on GT, AR and FERET databases, and the proposed method outperforms other related methods with variations on facial expressions, poses and lighting conditions, and it is especially superior in aging effect with the performance doubled. © 2012 ICIC International.
Showing items related by title, author, creator and subject.
Li, Billy Y.L. (2013)One of the most important advantages of automatic human face recognition is its nonintrusiveness property. Face images can sometime be acquired without user's knowledge or explicit cooperation. However, face images acquired ...
Li, Billy; Liu, Wan-Quan; An, Senjian; Krishna, Aneesh (2014)In this paper, we consider the problem of robust face recognition using color information. In this context, sparse representation-based algorithms are the state-of-the-art solutions for gray facial images. We will integrate ...
Li, Billy; Liu, Wan-Quan; An, Senjian; Krishna, Aneesh; Xu, T. (2012)Research on color face recognition in the existing literature is aimed to establish a color space that can have the most of the discriminative information from the original data. This mainly includes optimal combination ...