HiFlow: Training-free High-Resolution Image Generation with Flow-Aligned Guidance

Jiaqi Wang (Beijing University of Posts and Telecommunications) · Tong Wu (Beijing Institute for General Artificial Intelligence) · Dahua Lin (TTI Chicago) · Jiazi Bu (Shanghai Jiao Tong University) · Pengyang Ling (University of Science and Technology of China) · Yujie Zhou (Shanghai Jiaotong University) · Pan Zhang (Shanghai Artificial Intelligence Laboratory) · Xiaoyi Dong (University of Science and Technology of China) · Yuhang Zang (Shanghai Artificial Intelligence Laboratory) · Yuhang Cao (The Chinese University of Hong Kong)

acceleration alignmentdetail fidelitydiffusion modelsdirection alignmentflow-aligned guidancehigh-resolution synthesisinitialization alignmentintermediate stateslow-resolution samplingpersonalized variantspre-trained modelsstructure preservationtext-to-imagetraining-free strategiesvirtual reference flow

Text-to-image (T2I) diffusion/flow models have drawn considerable attention recently due to their remarkable ability to deliver flexible visual creations. Still, high-resolution image synthesis presents formidable challenges due to the scarcity and complexity of high-resolution content. Recent approaches have investigated training-free strategies to enable high-resolution image synthesis with pre-trained models. However, these techniques often struggle with generating high-quality visuals and tend to exhibit artifacts or low-fidelity details, as they typically rely solely on the endpoint of the low-resolution sampling trajectory while neglecting intermediate states that are critical for preserving structure and synthesizing finer detail. To this end, we present HiFlow, a training-free and model-agnostic framework to unlock the resolution potential of pre-trained flow models. Specifically, HiFlow establishes a virtual reference flow within the high-resolution space that effectively captures the characteristics of low-resolution flow information, offering guidance for high-resolution generation through three key aspects: initialization alignment for low-frequency consistency, direction alignment for structure preservation, and acceleration alignment for detail fidelity. By leveraging such flow-aligned guidance, HiFlow substantially elevates the quality of high-resolution image synthesis of T2I models and demonstrates versatility across their personalized variants. Extensive experiments validate HiFlow's capability in achieving superior high-resolution image quality over state-of-the-art methods.