SPAST: Arbitrary Style Transfer with Style Priors via Pre-trained Large-scale Model
This addresses the need for efficient and high-quality style transfer in image processing applications, though it appears incremental as it builds on pre-trained large-scale models.
The paper tackles the problem of arbitrary style transfer, where existing methods either produce low-quality images or are slow, by proposing SPAST, which generates high-quality stylized images with reduced inference time compared to state-of-the-art methods.
Given an arbitrary content and style image, arbitrary style transfer aims to render a new stylized image which preserves the content image's structure and possesses the style image's style. Existing arbitrary style transfer methods are based on either small models or pre-trained large-scale models. The small model-based methods fail to generate high-quality stylized images, bringing artifacts and disharmonious patterns. The pre-trained large-scale model-based methods can generate high-quality stylized images but struggle to preserve the content structure and cost long inference time. To this end, we propose a new framework, called SPAST, to generate high-quality stylized images with less inference time. Specifically, we design a novel Local-global Window Size Stylization Module (LGWSSM)tofuse style features into content features. Besides, we introduce a novel style prior loss, which can dig out the style priors from a pre-trained large-scale model into the SPAST and motivate the SPAST to generate high-quality stylized images with short inference time.We conduct abundant experiments to verify that our proposed method can generate high-quality stylized images and less inference time compared with the SOTA arbitrary style transfer methods.