Image conversion method based on variation automatic encoder and generative adversarial network

Abstract

The invention provides an image conversion method based on a variation automatic encoder and the generative adversarial network. The method is mainly characterized by comprising the variation automatic encoder (VAE), weight sharing, generating the generative adversarial network (GAN) and learning, in the process, a non-monitored image is utilized to learn a bidirectional conversion function between two image domains in an image conversion network framework (UNIT), VAE and VAE are comprised, modeling for each image domain is carried out through utilizing the VAE and the VAE, mutual action of an adversarial training target and a weight sharing constraint is carried out, corresponding images are generated in the two image domains, the conversion image is associated with an input image of each domain, and image reconstruction flow and image conversion flow problems can be solved through training network combination. The method is advantaged in that the non-monitoring image is utilized to the image conversion framework, images in the two domains having not any relations are made to accomplish conversion, a corresponding training data set formed by the images is not needed, efficiency and practicality are improved, and the method can be developed to non-monitoring language conversion.

Description

technical field [0001] The invention relates to the field of image conversion, in particular to an image conversion method based on a variational automatic encoder and a generation confrontation network. Background technique [0002] With the emergence of photographic technology, the popularization of television and movies, advertisements, newspapers, books, the Internet and many other image-based social media are developing rapidly, image-related technologies are becoming more and more important. Mapping images from one domain to another (image-to-image translation) has a wide range of applications. For example, the marine environment has poor visibility and often has haze, etc. Predicting the relevant conditions of the environment through image conversion will facilitate maritime traffic management, fishing and other marine operations; realize the conversion of images from day to night, from sunny to rainy Transformation, etc., so that people can predict the environment i...

AI Technical Summary

Problems solved by technology

However, most of the existing image-to-image translation methods are based on supervised learning, requiring a training dataset consisting of two corresponding image pairs, but it is difficult to obtain the required image pairs for many applications, thus hindering the translation. inconvenient

Images