One Diffusion Step to Real-World Super-Resolution via Flow Trajectory Distillation

TL;DR

FluxSR introduces a flow trajectory distillation approach to enable one-step diffusion-based real-world image super-resolution, significantly reducing computational cost while enhancing image quality and realism.

Contribution

The paper proposes FluxSR, a novel one-step diffusion method using flow trajectory distillation and new loss functions to improve real-world super-resolution performance.

Findings

Outperforms existing one-step diffusion-based Real-ISR methods

Reduces inference latency by using a single diffusion step

Addresses high-frequency artifacts with novel loss functions

Abstract

Diffusion models (DMs) have significantly advanced the development of real-world image super-resolution (Real-ISR), but the computational cost of multi-step diffusion models limits their application. One-step diffusion models generate high-quality images in a one sampling step, greatly reducing computational overhead and inference latency. However, most existing one-step diffusion methods are constrained by the performance of the teacher model, where poor teacher performance results in image artifacts. To address this limitation, we propose FluxSR, a novel one-step diffusion Real-ISR technique based on flow matching models. We use the state-of-the-art diffusion model FLUX.1-dev as both the teacher model and the base model. First, we introduce Flow Trajectory Distillation (FTD) to distill a multi-step flow matching model into a one-step Real-ISR. Second, to improve image realism and address high-frequency artifact issues in generated images, we propose TV-LPIPS as a perceptual loss and introduce Attention Diversification Loss (ADL) as a regularization term to reduce token similarity in transformer, thereby eliminating high-frequency artifacts. Comprehensive experiments demonstrate that our method outperforms existing one-step diffusion-based Real-ISR methods. The code and model will be released at https://github.com/JianzeLi-114/FluxSR.