Information generation method and device

Abstract

The invention discloses an information generation method and device. The information generation method comprises steps of obtaining an image to be detected and human face area information which is obtained through performing human face detection on the image to be detected in advance and is used for indicating a human face area in the image to be detected, performing enlargement on the human facearea indicated by the human face area information, taking the enlarged area as a target area, capturing an image in the target area, inputting the captured image into a pre-trained convolution neuralnetwork to obtain image characteristic information, wherein the convolution neural network is used for extracting image characteristics, performing analysis on image characteristic information and determining positions of key points of each human face and a probability of not shielding each key point of each human face in the target area. The information generation method and device improve richness of information generation.

Description

technical field [0001] The present application relates to the field of computer technology, specifically to the field of Internet technology, and in particular to an information generation method and device. Background technique [0002] With the development of Internet technology, face detection technology has been applied to more and more fields. For example, identity verification and the like can be performed through face detection. Usually, it is necessary to locate the key points of the face in the face image (such as corners of eyes, corners of the mouth, points in the contour, etc.) to generate position information of the key points of the face. [0003] However, when the face wears sunglasses, masks, etc., some key points of the face in the face image are occluded. The existing methods only support the generation of position information of key points of the face, and cannot generate judgment information on whether the key points of the face are occluded, which is n...

AI Technical Summary

Problems solved by technology

[0003] However, when the face wears sunglasses, masks, etc., some key points of the face in the face image are blocked

The existing methods only support the generation of position information of key points of the face, and cannot generate judgment information on whether the key points of the face are occluded, which is not conducive to further image processing (such as occlusion judgment, accessories judgment, face quality detection, etc. )

Therefore, the existing methods have the problem of low richness of information generation

Images