Face search method and system

Abstract

The invention relates to a face search method and system. The method includes the following steps that: a network framework for face retrieval is constructed, and the network framework is optimized; parameters in the network framework are adjusted according to a reverse propagation mode; training sample pictures are inputted into the adjusted network framework, so that test cases can be generated; quantization calculation is performed on the test cases according to a sign function, so that the binary codes of the test cases are obtained, and Hamming distances between the training sample pictures are calculated through the binary codes; and the approximation degrees of the training samples are sequenced according to the Hamming distances, and therefore, the training of the network framework for face retrieval is completed; and face pictures to be retrieved are inputted into the trained network framework for face retrieval, so that the face pictures to be retrieved can be retrieved, and the approximation degree-sequenced face pictures to be retrieved can be obtained. According to the face search method and system of the invention, the network and outputted features are optimized from the above aspects, so that the same accuracy can be maintained under a large-scale face database or the retrieval of face images can be performed quickly with accuracy decreased in a range as small as possible.

Description

technical field [0001] The invention relates to the technical field of image face retrieval, in particular to a face retrieval method and system. Background technique [0002] With the continuous improvement and improvement of face recognition methods, high-recognition face recognition methods are constantly being used in various fields, and many search engines that follow have also continuously launched face retrieval functions, such as Baidu image recognition, etc., regardless of Whether it is this type of search engine or the search database of the social security department, etc., are constantly expanding and updating. Many databases have reached a million-level scale, and searching a face picture or a person's face picture in such a large-scale database Only using linear sorting for comparison and identification will cause a very large calculation cost, thereby greatly reducing retrieval efficiency. Then, if a fast and accurate retrieval method can be developed in a mi...

AI Technical Summary

Problems solved by technology

Dimensionality reduction outside the network, that is, extracting high-dimensional features from the network, and using appropriate dimensionality reduction methods such as PCA to perform feature dimensionality

Images