Unified image restoration and enhancement: Degradation calibrated cycle reconstruction diffusion model

Xue, Minglong; He, Jinhong; Palaiahnakote, Shivakumara; Zhou, Mingliang

Abstract

Image restoration and enhancement are pivotal for numerous computer vision appli-
cations, yet unifying these tasks efficiently remains a significant challenge. Inspired
by the iterative refinement capabilities of diffusion models, we propose CycleRDM, a
novel framework designed to unify restoration and enhancement tasks while achiev-
ing high-quality mapping. Specifically, CycleRDM first learns the mapping relation-
ships among the degraded domain, the rough normal domain, and the normal domain
through a two-stage diffusion inference process. Subsequently, we transfer the final
calibration process to the wavelet low-frequency domain using discrete wavelet trans-
form, performing fine-grained calibration from a frequency domain perspective by
leveraging task-specific frequency spaces. To improve restoration quality, we design
a feature gain module for the decomposed wavelet high-frequency domain to elim-
inate redundant features. Additionally, we employ multimodal textual prompts and
Fourier transform to drive stable denoising and reduce randomness during the infer-
ence process. After extensive validation, CycleRDM can be effectively generalized
to a wide range of image restoration and enhancement tasks while requiring only a
small number of training samples to be significantly superior on various benchmarks
of reconstruction quality and perceptual quality. The source code will be available at
https://github.com/hejh8/CycleRDM.

• Journal Article Type: Article
• Acceptance Date: Jun 28, 2025
• Publication Date: 2026-03
• Deposit Date: Jul 20, 2025
• Publicly Available Date: Apr 1, 2028
• Journal: Pattern Recognition
• Print ISSN: 0031-3203
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• DOI: https://doi.org/10.1016/j.patcog.2025.112073