An Effective Unconstrained Correlation Filter and Its Kernelization for Face Recognition
This work addresses face recognition challenges by enhancing correlation filter methods, but it appears incremental as it builds on existing Class-dependence Feature Analysis (CFA) techniques.
The paper tackled the problem of robust face recognition by proposing an unconstrained correlation filter (UOOTF) that removes hard constraints on origin correlation outputs, and extended it with a kernel technique for non-linear separability, resulting in improved performance for unseen patterns.
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.