Face Image Resolution versus Face Recognition Performance Based on Two Global Methods

Face recognition is an interesting topic in computer vision and object recognition. Researchers have proposed numeric recognition methods under the various conditions such as different pose, illumination and expression. However no research has been done to depict the relation between face image resolution and face recognition performance so far. In this paper, we divide the face image information into the discriminative and structure information, and conjecture that the face recognition rate will level off when face image resolution arrives at one certain resolution threshold. Then, we propose several PCA and LDA based face recognition experiments and present statistical results to depict how face recognition rate changes with face image resolution. The experimental results validate the conjecture. Moreover, we propose several experiments to demonstrate that only enhancing the structure information could not help to improve the recognition performance.