QwenStyle: Content-Preserving Style Transfer with Qwen-Image-Edit
This work addresses style transfer for image editing applications, but it appears incremental as it builds on existing Qwen-Image-Edit capabilities.
The paper tackles the challenge of content-preserving style transfer in Diffusion Transformers by proposing QwenStyle, a model trained on Qwen-Image-Edit with a Curriculum Continual Learning framework, achieving state-of-the-art performance in style similarity, content consistency, and aesthetic quality.
Content-Preserving Style transfer, given content and style references, remains challenging for Diffusion Transformers (DiTs) due to its internal entangled content and style features. In this technical report, we propose the first content-preserving style transfer model trained on Qwen-Image-Edit, which activates Qwen-Image-Edit's strong content preservation and style customization capability. We collected and filtered high quality data of limited specific styles and synthesized triplets with thousands categories of style images in-the-wild. We introduce the Curriculum Continual Learning framework to train QwenStyle with such mixture of clean and noisy triplets, which enables QwenStyle to generalize to unseen styles without degradation of the precise content preservation capability. Our QwenStyle V1 achieves state-of-the-art performance in three core metrics: style similarity, content consistency, and aesthetic quality.