Abstract
In the image acquisition process, various forms of degradation, includingnoise, haze, and rain, are frequently introduced. These degradations typicallyarise from the inherent limitations of cameras or unfavorable ambientconditions. To recover clean images from degraded versions, numerousspecialized restoration methods have been developed, each targeting a specifictype of degradation. Recently, all-in-one algorithms have garnered significantattention by addressing different types of degradations within a single modelwithout requiring prior information of the input degradation type. However,these methods purely operate in the spatial domain and do not delve into thedistinct frequency variations inherent to different degradation types. Toaddress this gap, we propose an adaptive all-in-one image restoration networkbased on frequency mining and modulation. Our approach is motivated by theobservation that different degradation types impact the image content ondifferent frequency subbands, thereby requiring different treatments for eachrestoration task. Specifically, we first mine low- and high-frequencyinformation from the input features, guided by the adaptively decoupled spectraof the degraded image. The extracted features are then modulated by abidirectional operator to facilitate interactions between different frequencycomponents. Finally, the modulated features are merged into the original inputfor a progressively guided restoration. With this approach, the model achievesadaptive reconstruction by accentuating the informative frequency subbandsaccording to different input degradations. Extensive experiments demonstratethat the proposed method achieves state-of-the-art performance on differentimage restoration tasks, including denoising, dehazing, deraining, motiondeblurring, and low-light image enhancement. Our code is available athttps://github.com/c-yn/AdaIR.