SPG: Style-Prompting Guidance for Style-Specific Content Creation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liang, Qian, Chen, Zichong, Zhou, Yang, Huang, Hui
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916901418958848
author Liang, Qian
Chen, Zichong
Zhou, Yang
Huang, Hui
author_facet Liang, Qian
Chen, Zichong
Zhou, Yang
Huang, Hui
contents Although recent text-to-image (T2I) diffusion models excel at aligning generated images with textual prompts, controlling the visual style of the output remains a challenging task. In this work, we propose Style-Prompting Guidance (SPG), a novel sampling strategy for style-specific image generation. SPG constructs a style noise vector and leverages its directional deviation from unconditional noise to guide the diffusion process toward the target style distribution. By integrating SPG with Classifier-Free Guidance (CFG), our method achieves both semantic fidelity and style consistency. SPG is simple, robust, and compatible with controllable frameworks like ControlNet and IPAdapter, making it practical and widely applicable. Extensive experiments demonstrate the effectiveness and generality of our approach compared to state-of-the-art methods. Code is available at https://github.com/Rumbling281441/SPG.
format Preprint
id arxiv_https___arxiv_org_abs_2508_11476
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SPG: Style-Prompting Guidance for Style-Specific Content Creation
Liang, Qian
Chen, Zichong
Zhou, Yang
Huang, Hui
Graphics
Computer Vision and Pattern Recognition
Although recent text-to-image (T2I) diffusion models excel at aligning generated images with textual prompts, controlling the visual style of the output remains a challenging task. In this work, we propose Style-Prompting Guidance (SPG), a novel sampling strategy for style-specific image generation. SPG constructs a style noise vector and leverages its directional deviation from unconditional noise to guide the diffusion process toward the target style distribution. By integrating SPG with Classifier-Free Guidance (CFG), our method achieves both semantic fidelity and style consistency. SPG is simple, robust, and compatible with controllable frameworks like ControlNet and IPAdapter, making it practical and widely applicable. Extensive experiments demonstrate the effectiveness and generality of our approach compared to state-of-the-art methods. Code is available at https://github.com/Rumbling281441/SPG.
title SPG: Style-Prompting Guidance for Style-Specific Content Creation
topic Graphics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.11476