Image convolutional neural network style migration method based on local and global optimization fusion

Abstract

The invention discloses an image convolutional neural network style migration method based on local and global optimization fusion, and the method comprises the steps: firstly, selecting a content image and a style image which need to be converted, and employing a deep convolutional neural network VGG-19 as an image advanced style feature extraction basic model; selecting local optimization content constraint layer and a style constraint layer from the VGG-19 model, establishing a new network model F1, and defining an image style migration loss function based on local optimization; then selecting a global optimization style constraint layer from the VGG-19 model, establishing a new network model F2, and then performing linear fusion on the image style migration local optimization loss function and the image style migration global optimization loss function to obtain a total loss function; finally, initializing a noise image, respectively extracting the content image, the style image and the feature image by adopting the network models F1 and F2 according to the image style migration loss function, minimizing the loss function through multiple iterations by utilizing a gradient descent method, and finally generating a style migration result image.

Description

technical field [0001] The invention relates to the technical field of deep learning, in particular to a convolutional neural network style transfer method based on local and global optimization fusion images. Background technique [0002] Transferring the style of one image to another is considered an image texture style transfer problem. In image texture style transfer, the goal is to use one image (style image) to provide texture style, and another image to provide content (content image), and the synthesized image is required to have the texture style of the style image, but the content image needs to be preserved. semantic content. [0003] For texture synthesis, there are a large number of powerful non-parametric methods, in which the process of using CNN to fuse the semantic content of a picture with different styles is called neural style transfer (Neural Style Transfer), Gatys et al. in CVPR2016 (International An oral report article "image Style Transfer Using Con...

AI Technical Summary

Problems solved by technology

However, they can also produce serious errors when there are partial matching errors

Compared with methods based on local feature optimization methods, global feature optimization methods can better preserve the structure and color of style images, while detailed styles may not be fully captured

Images