Artistic Neural Style Transfer Algorithms with Activation Smoothing
Computer Vision and Pattern Recognition
2024-11-13 v1 Image and Video Processing
Abstract
The works of Gatys et al. demonstrated the capability of Convolutional Neural Networks (CNNs) in creating artistic style images. This process of transferring content images in different styles is called Neural Style Transfer (NST). In this paper, we re-implement image-based NST, fast NST, and arbitrary NST. We also explore to utilize ResNet with activation smoothing in NST. Extensive experimental results demonstrate that smoothing transformation can greatly improve the quality of stylization results.
Cite
@article{arxiv.2411.08014,
title = {Artistic Neural Style Transfer Algorithms with Activation Smoothing},
author = {Xiangtian Li and Han Cao and Zhaoyang Zhang and Jiacheng Hu and Yuhui Jin and Zihao Zhao},
journal= {arXiv preprint arXiv:2411.08014},
year = {2024}
}
Comments
8 pages,7 figures