A Gabor-Block-Based Kernel Discriminative Common Vector Approach Using Cosine Kernels for Human Face Recognition

Figure 5

Face recognition performance of the proposed method using cosine kernel function and considering the first 3 images as training set on the FERET dataset, with the three different similarity measures: cos (cosine similarity measure), 𝐿₂ (𝐿₂ distance measure), 𝐿₁ (𝐿₁ distance measure).