Artistic Neural Style Transfer Algorithms with Activation Smoothing
X T Li, Han Cao, Zhaoyang Zhang, Jiacheng Hu, Yuhui Jin, Zihao Zhao · 2025
The works of Gatys et al. [1], [2], demonstrated the capability of Convolutional Neural Networks 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. Through a wide range of experiments, it has been verified that employing smoothing transformations plays a crucial role in boosting the visual fidelity of stylized images.