Face recognition under pose variations

In recent years, face recognition has attracted significant attention from the research and commercial communities. Because of the wide variation in face images, face recognition for real applications remains a very challenging problem. A large number of face recognition algorithms, along with their modifications, have been proposed over the past three decades. This paper presents a review of the typical algorithms that aim to overcome one of the main obstacles in the face recognition task, the variations in face pose. These algorithms are categorized and briefly described. Future research challenges in pose-invariant face recognition are also identified.