AddSR: Accelerating diffusion-based blind super-resolution with adversarial diffusion distillation

doi:10.1016/j.patcog.2025.113012

AddSR: Accelerating diffusion-based blind super-resolution with adversarial diffusion distillation

2026-01-01

1

PRE

AI

DOI：10.1016/j.patcog.2025.113012

Original

Original request for help

Save

Abstract

En 中文

Blind super-resolution methods based on Stable Diffusion (SD) demonstrate impressive generative capabilities in reconstructing clear, high-resolution (HR) images with intricate details from low-resolution (LR) inputs. However, their practical applicability is often limited by poor efficiency, as they typically require tens or even dozens of sampling steps. Inspired by Adversarial Diffusion Distillation (ADD), we incorporate this approach to design a highly effective and efficient blind super-resolution method. Nonetheless, two challenges arise: First, the original ADD significantly reduces result fidelity, leading to a perception-distortion imbalance. Second, SD-based methods are sensitive to the quality of the conditioning input, while LR images often have complex degradation, which further hinders effectiveness. To address these issues, we introduce a Timestep-Adaptive ADD (TA-ADD) to mitigate the perception-distortion imbalance caused by the original ADD. Furthermore, we propose a prediction-based self-refinement strategy to estimate HR, which allows for the provision of more high-frequency information without the need for additional modules. Extensive experiments show that our method, AddSR, generates superior restoration results while being significantly faster than previous SD-based state-of-the-art models (e.g., 7 x faster than SeeSR).

Keywords:

Super-resolution

Diffusion model

Generative prior

Efficiency

AddSR: Accelerating diffusion-based blind super-resolution with adversarial diffusion distillation

Abstract

Journal

Organization

Cited Papers