{"type":"ScholarlyArticle","title":"A Timestep-Adaptive Frequency-Enhancement Framework for Diffusion-based Image Super-Resolution","authors":[{"name":"Yueying Li","slug":"yueying-li","given":"Yueying","family":"Li","affiliation":""},{"name":"Hanbin Zhao","slug":"hanbin-zhao","given":"Hanbin","family":"Zhao","affiliation":""},{"name":"Jiaqing Zhou","slug":"jiaqing-zhou","given":"Jiaqing","family":"Zhou","affiliation":""},{"name":"Guozhi Xu","slug":"guozhi-xu","given":"Guozhi","family":"Xu","affiliation":""},{"name":"Tianlei Hu","slug":"tianlei-hu","given":"Tianlei","family":"Hu","affiliation":""},{"name":"Gang Chen","slug":"gang-chen","given":"Gang","family":"Chen","affiliation":""},{"name":"Haobo Wang","slug":"haobo-wang","given":"Haobo","family":"Wang","affiliation":""}],"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.","doi":"10.24963/ijcai.2025/168","url":"https://ijcai.org/proceedings/2025/168","pdf":"https://ijcai.org/proceedings/2025/0168.pdf","markdown":"https://ijcai.org/proceedings/2025/168.md","bibtex":"https://ijcai.org/proceedings/2025/bibtex/168","volume":{"slug":"2025","title":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2025","url":"https://ijcai.org/proceedings/2025/","year":2025,"location":"Montreal, Canada","dates":"16-22 August 2025","editors":"James Kwok","isbn":"978-1-956792-06-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2025"},"section":"Computer Vision","track":"Main Track","number":168,"pages":{"first":1503,"last":1511},"keywords":["Machine Learning: ML: Generative models","Computer Vision: CV: Image and video synthesis and generation"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}