Hashing is most popular technique which provides an efficient and accurate way to nearest neighbor search in large scale data. In large scale image retrieval data is represents in the form of semantic similarity presented in labeled pair of images. Thus unsupervised techniques are efficient to provide solution for these problems, supervised hashing technique is required to provide desired solution. In this paper a survey over these techniques is presented. A Multiview alignment based hashing technique is presented which uses regularized kernel nonnegative matrix factorization (RKNMF) to enhance the performance of the nearest neighbor search, A composite hashing for multiple information search is presented. There are some other techniques are also presented, which presents an overview over the hashing techniques used for large scale image search.
展开▼
