A brief review on techniques for recognizing images under varying poses

Face recognition has achieved immense popularity in various fields because of its robustness and accuracy. But pose variation is still a major obstacle to overcome for effective face recognition in an uncontrolled environment. A wide variety of face recognition algorithms have been proposed in the past. In this paper we exhibit a review of some of the common algorithms that expect to conquer on the fundamental impediments in face recognition, i.e., pose variation. An outline of the recognized systems in each of this classification is given and some of the advantages and disadvantages of the algorithms specified in that are inspected. The key contribution of this paper is that we have analyzed the latest state of art techniques in Karhunen-Loeve expansion and Model based methods. From this analysis, we have found that Karhunen-Loeve expansion and Model based methods are giving best results for FERET database giving 95% and CMU-PIE database giving 98.8% recognition rate respectively. We have also tabulated the results and observations of the algorithms mentioned on being tested on some of the renowned and universal databases. Furthermore, an overview of the benefits of face recognition systems and its applications in the real world has also been discussed in this paper.