SPAST: Arbitrary Style Transfer with Style Priors via Pre-trained Large-scale Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Zhanjie, Zhang, Quanwei, Luan, Junsheng, Yang, Mengyuan, Wang, Yun, Zhao, Lei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916735402115072
author Zhang, Zhanjie
Zhang, Quanwei
Luan, Junsheng
Yang, Mengyuan
Wang, Yun
Zhao, Lei
author_facet Zhang, Zhanjie
Zhang, Quanwei
Luan, Junsheng
Yang, Mengyuan
Wang, Yun
Zhao, Lei
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2505_08695
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SPAST: Arbitrary Style Transfer with Style Priors via Pre-trained Large-scale Model
Zhang, Zhanjie
Zhang, Quanwei
Luan, Junsheng
Yang, Mengyuan
Wang, Yun
Zhao, Lei
Computer Vision and Pattern Recognition
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.
title SPAST: Arbitrary Style Transfer with Style Priors via Pre-trained Large-scale Model
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.08695