Method of Image Style Transfer Based on Edge Detection
Hanmin Ye, Wenjie Liu, Shiming Huang · 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) · 2020
In order to overcome the problem of edge information loss in the process of neural network processing, a method of neural network style transfer based on edge detection is presented. The edge information of the content image is extracted, and the edge information image is processed in the neural network together with the content image and the style image to constrain the edge information of the content image. Compared with Gatys algorithm and markov random field neural network algorithm, the content image edge structure after image style transfer is successfully retained.