Face recognition with enhanced privacy protection

This paper presents a novel approach for face based biometric recognition. The proposed method is based on the sorted index numbers (SIN) of appearance based facial features. A new algorithm is introduced to measure the similarity between SIN vectors. Due to the non-invertibility of the transformation from the original features to the SIN vectors, the proposed method can preserve the privacy of the users. The effectiveness of the proposed method is tested on a large generic data set, which contains images from several well known face databases. Experimental results demonstrate that the proposed solution may improve the recognition accuracy in both identification and verification scenarios.