Neural Artistic Style Transfer with Conditional Adversaria

TL;DR

This paper introduces a novel neural style transfer method using a unidirectional-GAN that achieves style independence, smaller model size, and efficient training while maintaining semantic accuracy.

Contribution

The paper presents a unidirectional-GAN architecture with cyclic consistency for style-independent neural style transfer, reducing model size and training complexity.

Findings

Achieved style-independent neural style transfer.

Reduced model size compared to traditional methods.

Ensured semantic accuracy in generated images.

Abstract

A neural artistic style transformation (NST) model can modify the appearance of a simple image by adding the style of a famous image. Even though the transformed images do not look precisely like artworks by the same artist of the respective style images, the generated images are appealing. Generally, a trained NST model specialises in a style, and a single image represents that style. However, generating an image under a new style is a tedious process, which includes full model training. In this paper, we present two methods that step toward the style image independent neural style transfer model. In other words, the trained model could generate semantically accurate generated image under any content, style image input pair. Our novel contribution is a unidirectional-GAN model that ensures the Cyclic consistency by the model architecture.Furthermore, this leads to much smaller model size and an efficient training and validation phase.