DRB-GAN: A Dynamic ResBlock Generative Adversarial Network for Artistic Style Transfer

TL;DR

DRB-GAN introduces a novel dynamic ResBlock architecture with style-aware attention and a specialized discriminator, achieving superior artistic style transfer quality and efficiency in both arbitrary and collection styles.

Contribution

The paper presents a new DRB-GAN model with Dynamic ResBlocks and style collection discrimination, advancing artistic style transfer capabilities.

Findings

Outperforms state-of-the-art style transfer methods in quality

Enables both arbitrary and collection style transfer

Demonstrates high efficiency and visual quality

Abstract

The paper proposes a Dynamic ResBlock Generative Adversarial Network (DRB-GAN) for artistic style transfer. The style code is modeled as the shared parameters for Dynamic ResBlocks connecting both the style encoding network and the style transfer network. In the style encoding network, a style class-aware attention mechanism is used to attend the style feature representation for generating the style codes. In the style transfer network, multiple Dynamic ResBlocks are designed to integrate the style code and the extracted CNN semantic feature and then feed into the spatial window Layer-Instance Normalization (SW-LIN) decoder, which enables high-quality synthetic images with artistic style transfer. Moreover, the style collection conditional discriminator is designed to equip our DRB-GAN model with abilities for both arbitrary style transfer and collection style transfer during the training stage. No matter for arbitrary style transfer or collection style transfer, extensive experiments strongly demonstrate that our proposed DRB-GAN outperforms state-of-the-art methods and exhibits its superior performance in terms of visual quality and efficiency. Our source code is available at \color{magenta}{\url{https://github.com/xuwenju123/DRB-GAN}}.