Video face identifying method

Abstract

The invention discloses a video face identifying method which comprises the following steps of S1, carrying out face detection and trace on video to obtain face sequences, S2, screening the face sequences to obtain a typical face frame set, S3, optimizing the typical face frame set based on a front face generating technique and an image super resolution technique to obtain a reinforced typical face frame set, and S4, comparing the reinforced typical face frame set with a preset static face image matching base to identify or verify faces. Compared with an existing video face identifying method, the video face identifying method filters and compensates change of video face postures and resolutions through the reinforced typical face frame set. Thus, the robustness of video face identification is improved.

Description

technical field [0001] The invention relates to the technical field of video monitoring, in particular to a video face recognition method. Background technique [0002] With the popularity of video surveillance systems and increasing development needs, video-based face recognition has made great progress, and it plays a very important role in intelligent transportation, access control, information security, security, security and other security fields. [0003] A main application mode of the existing video face recognition method is video to static image recognition, the video to static image recognition uses face video as input, and realizes recognition or verification by comparing with a static image face database . [0004] One type of method for video recognition of static images is to use a static face recognition method for each frame of face images to identify, and finally fuse the recognition results of all frames according to probability scoring, distance judgment ...

AI Technical Summary

Problems solved by technology

[0012] The problem with CCA is that it ignores the separability of different types of face samples in the public space;

[0016] The problem with PLS is that PLS is unsupervised and does not make full use of the identification information of c-type face samples.

[0017] The problems existing in the existing mapping learning methods are: the mapping matrix learned under the complex changes of the pose and resolution of the video face is inaccurate, and the identification information of the face category is not fully introduced, so the identification of the face recognizer performance is not strong enough

Images