A semi-supervised multi-modal multi-category image translation method

Abstract

The invention discloses a semi-supervised multi-modal multi-category image translation method, comprising steps: S1, input two images from different domains and a small number of labels; S2, send the input images and labels to an encoder, and encode The encoder is divided into a content encoder and a style encoder, using decoupling representation learning to decouple the image from the style encoder and the content encoder to obtain style encoding and content encoding respectively; S3, input the style encoding into the adversarial autoencoder , to complete image multi-category training; input the content code into the content confrontation learning network to complete image multi-modal transformation training; S4, realize image reconstruction and multi-modal transformation by splicing style coding and content coding. The invention solves the dilemma caused by the requirement of image translation diversity, and can generate multi-modal and multi-category cross-domain images through joint decoding of latent content coding and style coding.

Description

technical field [0001] The present invention relates to the technical fields of computer vision, computer graphics and machine learning, in particular to a semi-supervised, multi-modal and multi-category image translation method. Background technique [0002] With the continuous development of deep learning technology and image generation technology, a large number of excellent works have emerged in the field of image translation. Semi-supervised multi-modal and multi-category image translation is an important and challenging research problem in the field of computer vision. Semi-supervised multi-modal Stateful multi-category image translation has obvious application value in industrial design and other fields, and can be applied to various aspects such as image colorization, super-resolution generation, and style transfer. At present, the existing image translation is to transform the problem into a one-to-one image mapping, which needs to be clearly given two different ima...

AI Technical Summary

Problems solved by technology

At present, the existing image translation is to transform the problem into a one-to-one image mapping, which needs to be clearly given two different image domains, and in many scenarios, cross-domain image translation is multimodal, so , the existing cross-domain data translation cannot meet these needs

Images