In this project, we investigate and implement a promising search based face annotation scheme by mining large amount of weakly labelled facial images freely available on the World Wide Web (WWW). We proposes a review on various techniques used for detection and analysis of each technique and efficient-based approximation algorithm for large-scale label refinement problem. The face annotation has many real world applications. Most of the users use person’s name as the search query. So it is effective to label the images with their exact names. The automatic face recognition techniques can annotate the faces with exact labels and it also help to improve the search more efficiently annotations from them.  The challenging part of search based face annotation task is management of most familiar facial images and their weak labels. Different techniques are used in retrieving facial images based on search query. The efficiency and performance of annotating systems are improved tremendously by using edge detection technique. This approach is beneficial to find the correct data related to facial features. Firstly, at least one accurate keyword is required to enable text-based search for a set of semantically similar images. Then content-based search is performed on this set to retrieve visually similar images. At last, annotations are mined from the descriptions. Specifically, the task of face-name association should obey the following three constraints: a face can only be assigned to a name appearing in its associated caption or to null; a name can be assigned to at most one face; and a face can be assigned to at most one name. Many conventional methods have been proposed to tackle this task while suffering from some common. The face annotation technique with edge detection algorithm is efficient to  tackle problem in clustering based approximation algorithm and ULR algorithms can significantly boost the performance of the promising search based face annotation (SBFA) scheme.