---
title: "A Timestep-Adaptive Frequency-Enhancement Framework for Diffusion-based Image Super-Resolution"
authors:
  - "Yueying Li"
  - "Hanbin Zhao"
  - "Jiaqing Zhou"
  - "Guozhi Xu"
  - "Tianlei Hu"
  - "Gang Chen"
  - "Haobo Wang"
doi: "10.24963/ijcai.2025/168"
url: "https://ijcai.org/proceedings/2025/168"
pdf: "https://ijcai.org/proceedings/2025/0168.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "1503-1511"
track: "Main Track"
keywords:
  - "Machine Learning: ML: Generative models"
  - "Computer Vision: CV: Image and video synthesis and generation"
---

# A Timestep-Adaptive Frequency-Enhancement Framework for Diffusion-based Image Super-Resolution

Yueying Li, Hanbin Zhao, Jiaqing Zhou, Guozhi Xu, Tianlei Hu, Gang Chen, Haobo Wang

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 1503-1511. DOI: [10.24963/ijcai.2025/168](https://doi.org/10.24963/ijcai.2025/168)

## Abstract

Image super-resolution (ISR) is a classic and challenging problem in computer vision because of complex and unknown degradation patterns in the data collection process. Leveraging powerful generative priors, diffusion-based methods have recently established new state-of-the-art ISR performance, but their characteristics in the frequency domain are still underexplored. In this paper, we innovatively investigate their frequency-domain behaviors from a sampling timestep perspective. Experimentally, we find that current diffusion-based ISR algorithms exhibit insufficiency in different frequency components in distinct groups of timesteps during the sampling. To address this, we first propose a Timestep Division Controller that is able to adaptively divide the timesteps into groups based on the performance gradient across different components. Next, we design two dedicated modules --- the Amplitude and Phase Enhancement Module (APEM) and the High- and Low-Frequency Enhancement Module (HLEM), to regulate the information flow of distinct frequency-domain features. By adaptively enhancing specific frequency components at different stages of the sampling process, the two modules effectively compensate for the insufficient frequency-domain perception of diffusion-based ISR models. Extensive experiments on three benchmark datasets verify the superior ISR performance of our method, e.g., achieving an average 5.40% improvement on CLIP-IQA compared to the best diffusion-based ISR baseline.