Sissi: Zero-shot Style-guided Image Synthesis via Semantic-style Integration

TL;DR

Sissi introduces a training-free, semantic-style integration method for zero-shot style-guided image synthesis that balances content and style fidelity using multimodal attention and dynamic reweighting.

Contribution

The paper presents a novel in-context learning framework for style-guided image synthesis that avoids retraining and improves style-content balance through dynamic attention reweighting.

Findings

Achieves high-fidelity stylization with balanced semantic and style adherence.

Outperforms prior methods in visual quality and coherence.

Operates without task-specific retraining or expensive inversion procedures.

Abstract

Text-guided image generation has advanced rapidly with large-scale diffusion models, yet achieving precise stylization with visual exemplars remains difficult. Existing approaches often depend on task-specific retraining or expensive inversion procedures, which can compromise content integrity, reduce style fidelity, and lead to an unsatisfactory trade-off between semantic prompt adherence and style alignment. In this work, we introduce a training-free framework that reformulates style-guided synthesis as an in-context learning task. Guided by textual semantic prompts, our method concatenates a reference style image with a masked target image, leveraging a pretrained ReFlow-based inpainting model to seamlessly integrate semantic content with the desired style through multimodal attention fusion. We further analyze the imbalance and noise sensitivity inherent in multimodal attention fusion and propose a Dynamic Semantic-Style Integration (DSSI) mechanism that reweights attention between textual semantic and style visual tokens, effectively resolving guidance conflicts and enhancing output coherence. Experiments show that our approach achieves high-fidelity stylization with superior semantic-style balance and visual quality, offering a simple yet powerful alternative to complex, artifact-prone prior methods.