arXiv:2404.15743v3 Announce Type: replace 
Abstract: Recent style transfer problems are still largely dominated by Generative Adversarial Network (GAN) from the perspective of cross-domain image-to-image (I2I) translation, where the pivotal issue is to learn and transfer target-domain style patterns onto source-domain content images. This paper handles the problem of translating real pictures into traditional Chinese ink-wash paintings, i.e., Chinese ink-wash painting style transfer. Though a wide range of I2I models tackle this problem, a notable challenge is that the content details of the source image could be easily erased or corrupted due to the transfer of ink-wash style elements. To remedy this issue, we propose to incorporate saliency detection into the unpaired I2I framework to regularize image content, where the detected saliency map is utilized from two aspects: (\romannumeral1) we propose saliency IOU (SIOU) loss to explicitly regularize object content structure by enforcing saliency consistency before and after image stylization; (\romannumeral2) we propose saliency adaptive normalization (SANorm) which implicitly enhances object structure integrity of the generated paintings by dynamically injecting image saliency information into the generator to guide stylization process. Besides, we also propose saliency attended discriminator which harnesses image saliency information to focus generative adversarial attention onto the drawn objects, contributing to generating more vivid and delicate brush strokes and ink-wash textures. Extensive qualitative and quantitative experiments demonstrate superiority of our approach over related advanced image stylization methods in both GAN and diffusion model paradigms.

تقدم ورقة بحثية حديثة SRAGAN، وهو نهج جديد لنقل الأسلوب يستخدم الشبكات التنافسية التوليدية (GAN) لتحويل الصور الحقيقية إلى لوحات تقليدية صينية بالحبر. هذه الخطوة مهمة لأنها لا تعزز فقط التمثيل الفني للصور، بل تربط أيضًا التكنولوجيا الحديثة بأشكال الفن الكلاسيكية، مما يظهر إمكانيات الذكاء الاصطناعي في المجالات الإبداعية.

Un artículo reciente presenta SRAGAN, un enfoque novedoso para la transferencia de estilo que utiliza Redes Generativas Antagónicas (GAN) para transformar imágenes reales en pinturas tradicionales chinas de tinta. Este avance es significativo ya que no solo mejora la representación artística de las imágenes, sino que también une la tecnología moderna con las formas de arte clásicas, mostrando el potencial de la IA en los campos creativos.

Un récent article présente SRAGAN, une nouvelle approche du transfert de style qui utilise des réseaux antagonistes génératifs (GAN) pour transformer des images réelles en peintures à l'encre de Chine traditionnelles. Cette avancée est importante car elle améliore non seulement la représentation artistique des images, mais elle relie également la technologie moderne aux formes d'art classiques, montrant le potentiel de l'IA dans les domaines créatifs.

A recent paper introduces SRAGAN, a novel approach to style transfer that utilizes Generative Adversarial Networks (GANs) to transform real images into traditional Chinese ink-wash paintings. This advancement is significant as it not only enhances the artistic representation of images but also bridges modern technology with classical art forms, showcasing the potential of AI in creative fields.

SRAGAN: Saliency Regularized and Attended Generative Adversarial Network for Chinese Ink-wash Painting Style Transfer

Was this article worth reading? Share it

Canny Valley

Golan AI

TattooGen