Image conversion model training method and device, heterogeneous face recognition method and device, and equipment

Abstract

The invention provides an image conversion model training method and device, a heterogeneous face recognition method and device and equipment, and the method comprises the steps: obtaining a to-be-recognized face sketch, and cutting the to-be-recognized face sketch to obtain a face region sketch; inputting the face region sketch into a pre-trained image conversion model for processing to generatea second face synthesis image; performing feature extraction on the second face synthesis image to obtain a feature vector of the second face synthesis image; and matching the feature vector of the second face synthesis image with feature vectors of a plurality of real face images stored in a database to obtain a face recognition result. Therefore, the face area is cut from the to-be-recognized face sketch to reduce the interference of the background area on face recognition, and meanwhile, the preset image conversion model is adopted to perform image conversion to convert heterogeneous face recognition into homogeneous face recognition, so that the accuracy of heterogeneous face recognition is improved.

Description

technical field [0001] The present application relates to the technical fields of artificial intelligence and image processing, and in particular to an image conversion model training method, heterogeneous face recognition method, device and equipment. Background technique [0002] With the development of artificial intelligence, face recognition has always been a hot field of research in all walks of life, and significant progress has been made on the basis of deep convolutional neural networks, and it is widely used in all walks of life in society. For example, in the case investigation of the public security department, face recognition can be used to narrow down the scope of potential suspects and reduce the difficulty of investigation. However, in some cases, a clear and complete face image cannot be obtained, or even there is no face image record. At this time, a sketch expert is usually required to draw a sketch of the suspect’s face based on a low-resolution image or...

AI Technical Summary

Problems solved by technology

However, in some cases, a clear and complete face image cannot be obtained, or even there is no face image record. At this time, a sketch expert is usually required to draw a sketch of the suspect’s face based on a low-resolution image or an eyewitness description. Face recognition, however, due to the large differences between the face image collected by the image acquisition device and the sketch (such as shape, texture and color, etc.), it is difficult to use the traditional face recognition algorithm to identify the face of the police according to the sketch. Exact search in the database

[0003] In related technologies, the face recognition method for sketches is to use convolutional neural network to mine the potential relationship between sketches and real images, and to model based on the nonlinear mapping of the two, so as to convert sketches to real images , and then recognize the converted image; or use the generative confrontation network to generate an image with more realistic texture through the confrontation loss for recognition, but the accuracy rate is not very high

Images