A Oil Painting Style Migration Algorithm Based on Convolutional Neural Network

EasyChair Preprint 919

Abstract

The oil painting style is to add the sand painting style information of an image to any image, and maintain the semantic content of the image to produce a new ornamental image, mainly introducing a convolutional neural network based image. This paper mainly introduces an image style algorithm based on convolutional neural network, which can separate and reorganize the image content and style of natural pictures, and then realize the oil painting migration of pictures. The algorithm combines the content of any image with many well-known oil painting styles to achieve high-perceptual quality artwork. The experimental results show that the algorithm can achieve deep fusion of image content and optimization style, which proves the effectiveness of the algorithm.

Keyphrases: Convolutional Neural Network, image stylization, oil painting

@booklet{EasyChair:919,
  author    = {Jiang Min and Sheng Ran and Zhu De and Duan Yunsheng and Li Fangfang and Sun Dong},
  title     = {A Oil Painting Style Migration Algorithm Based on Convolutional Neural Network},
  howpublished = {EasyChair Preprint 919},
  year      = {EasyChair, 2019}}