Portrait image generation method and device, readable medium and electronic equipment

Abstract

The invention discloses a portrait image generation method and device, a readable medium and electronic equipment, and is applied to the field of image generation. By controlling the target noise vector input into the network unit of the generator, the feature quality of the generated portrait image library is qualified, the richness is sufficiently generalized, and the number is sufficiently sufficient. Specifically, the method comprises the following steps: acquiring a plurality of random noise vectors; carrying out nonlinear transformation on the plurality of random noise vectors to obtaina plurality of target noise vectors, wherein the plurality of random noise vectors are in one-to-one correspondence with the plurality of target noise vectors; according to the multiple target noise vectors, obtaining multiple sets of vector sequences, wherein the number of the target noise vectors contained in each set of vector sequence is n, the target noise vectors contained in different vector sequences are different, and n is a positive integer larger than or equal to 2; and generating a plurality of human image images based on the plurality of groups of vector sequences, wherein the plurality of groups of vector sequences are in one-to-one correspondence with the plurality of human image images. Specifically, the method is applied to a scene of generating a portrait image.

Description

technical field [0001] The present application relates to the field of image generation, in particular to a method, device, readable medium and electronic equipment for generating portrait images. Background technique [0002] With the application of face recognition technology in security, finance and other fields, the quality requirements for products related to face recognition technology have also become higher. Usually, the accuracy supported by the face recognition model used by the product and the size of the bottom library (ie, the portrait database) are used as important indicators to measure the quality of the product. For example, the use of portrait aggregation technology in the face recognition model is conducive to improving the accuracy of portrait recognition and increasing the size of the portrait library. Among them, portrait aggregation refers to using the face recognition model to calculate different photos of the same person according to the similarity ...

AI Technical Summary

Problems solved by technology

However, face data is protected by the government because it involves personal privacy, so it is difficult to obtain, resources are scarce and cannot form a market, which makes it difficult to obtain large-scale portrait databases

If it is considered to forge a virtual face for the test of the corresponding model of the portrait aggregation file, then due to the general technology in the image processing field such as pixel fusion and region cropping, it is impossible to form new face features when forging a virtual face, which will lead to the generation of virtual faces. Face cannot be used for portrait aggregation verification

Images