Silent living body detection method and device, computer equipment and storage medium

Abstract

The embodiment of the invention discloses a silent living body detection method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a to-be-detected video; performing optical flow feature extraction on the to-be-detected video to obtain a human face optical flow feature map and an RGB face image; inputting the face optical flow feature map and the RGB face image into a silent living body detection model for silent living body detection to obtain a detection result; wherein the silent living body detection model is obtained by training a deep learning model by taking a human face optical flow feature map with a living body or non-living body label and an RGB human face image as a sample set. By implementing the method provided by the embodiment of the invention, the living body detection accuracy can be improved, and silent living body detection in a real sense is realized.

Description

technical field [0001] The present invention relates to a living body detection method, more specifically to a silent living body detection method, device, computer equipment and storage medium. Background technique [0002] At present, the most effective silent liveness detection method is the face liveness detection method based on the deep learning method. In order to effectively judge the subtle differences between living and non-living bodies in RGB face images due to light reflection, and make the corresponding algorithm have a lot of For good generalization ability, the methods currently used include the following three methods. One is to input continuous multi-frame face images, and design features by capturing the difference between living and non-living micro-movements; the second is to use the central difference volume Actively capture the subtle differences in the edge information of the face; there is also a multi-modal fusion face live detection method, which u...

AI Technical Summary

Problems solved by technology

However, the above-mentioned methods are relatively complicated, and multi-frame face image input requires or image storage space. In addition to complexity, all face liveness detection methods have a disadvantage that the generalization ability is poor, and cross-camera and cross-lighting environments are different. There is no good solution, and it is greatly affected by the light. When the face is not moving, the optical flow movement characteristics are not obvious, which may easily lead to low accuracy of liveness detection, and it is impossible to achieve true silent liveness detection.

Images