SplitFlux: Learning to Decouple Content and Style from a Single Image

TL;DR

SplitFlux introduces a novel method for disentangling content and style in images by fine-tuning single stream blocks, leading to improved content preservation and stylization quality in image generation.

Contribution

The paper proposes SplitFlux, a new approach that effectively separates content and style using LoRA fine-tuning and novel components like Rank-Constrained Adaptation and Visual-Gated LoRA.

Findings

SplitFlux outperforms state-of-the-art methods in content preservation.

The method achieves superior stylization quality across diverse scenarios.

Extensive experiments validate the effectiveness of the proposed approach.

Abstract

Disentangling image content and style is essential for customized image generation. Existing SDXL-based methods struggle to achieve high-quality results, while the recently proposed Flux model fails to achieve effective content-style separation due to its underexplored characteristics. To address these challenges, we conduct a systematic analysis of Flux and make two key observations: (1) Single Stream Blocks are essential for image generation; and (2) Early single stream blocks mainly control content, whereas later blocks govern style. Based on these insights, we propose SplitFlux, which disentangles content and style by fine-tuning the single stream blocks via LoRA, enabling the disentangled content to be re-embedded into new contexts. It includes two key components: (1) Rank-Constrained Adaptation. To preserve content identity and structure, we compress the rank and amplify the magnitude of updates within specific blocks, preventing content leakage into style blocks. (2) Visual-Gated LoRA. We split the content LoRA into two branches with different ranks, guided by image saliency. The high-rank branch preserves primary subject information, while the low-rank branch encodes residual details, mitigating content overfitting and enabling seamless re-embedding. Extensive experiments demonstrate that SplitFlux consistently outperforms state-of-the-art methods, achieving superior content preservation and stylization quality across diverse scenarios.