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

TL;DR

This paper introduces Style-Prompting Guidance (SPG), a new sampling strategy for controlling visual style in text-to-image diffusion models, improving style consistency without sacrificing semantic accuracy.

Contribution

The paper proposes SPG, a novel style control method that guides diffusion models using style noise vectors and integrates seamlessly with existing guidance frameworks.

Findings

SPG enhances style consistency in generated images.

The method maintains high semantic fidelity.

It is compatible with various controllable diffusion frameworks.

Abstract

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.