Nearest neighbor search is a problem of finding the data points from thedatabase such that the distances from them to the query point are the smallest.Learning to hash is one of the major solutions to this problem and has beenwidely studied recently. In this paper, we present a comprehensive survey ofthe learning to hash algorithms, categorize them according to the manners ofpreserving the similarities into: pairwise similarity preserving, multiwisesimilarity preserving, implicit similarity preserving, as well as quantization,and discuss their relations. We separate quantization from pairwise similaritypreserving as the objective function is very different though quantization, aswe show, can be derived from preserving the pairwise similarities. In addition,we present the evaluation protocols, and the general performance analysis, andpoint out that the quantization algorithms perform superiorly in terms ofsearch accuracy, search time cost, and space cost. Finally, we introduce a fewemerging topics.
translated by 谷歌翻译