In this paper, an effective unconstrained correlation filter called Uncon- strained Optimal Origin Tradeoff Filter (UOOTF) is presented and applied to robust face recognition. Compared with the conventional correlation filters in Class-dependence Feature Analysis (CFA), UOOTF improves the overall performance for unseen patterns by removing the hard constraints on the origin correlation outputs during the filter design. To handle non-linearly separable distributions between different classes, we further develop a non- linear extension of UOOTF based on the kernel technique. The kernel ex- tension of UOOTF allows for higher flexibility of the decision boundary due to a wider range of non-linearity properties. Experimental results demon- strate the effectiveness of the proposed unconstrained correlation filter and its kernelization in the task of face recognition.
@article{arxiv.1603.07800,
title = {An Effective Unconstrained Correlation Filter and Its Kernelization for Face Recognition},
author = {Yan Yan and Hanzi Wang and Cuihua Li and Chenhui Yang and Bineng Zhong},
journal= {arXiv preprint arXiv:1603.07800},
year = {2016}
}