An Automatic Face Annotation System Featuring High Accuracy for Online Social Networks

The development of fully automatic face annotation techniques in online social networks (OSNs) is currently very important for effective management and organization of the large numbers of personal photos shared on social network platforms. In this paper, we construct the personalized and adaptive Fused Face Recognition unit for each member, which uses the Adaboost algorithm to fuse several different types of base classifiers to produce highly reliable face annotation results. The experiment results demonstrate that our proposed approach achieves a significantly higher level of efficacy, outperforming other state-of-the-art face annotation methods for real-life personal photos featuring pose variations. Our evaluation methodologies produced respective F-measure and Similarity accuracy rates that were 57.99% and 54.23% higher for the proposed method in comparison to other tested methods.