Degradation-Aware Feature Perturbation for All-in-One Image Restoration
This addresses the problem of task interference in multi-task image restoration for computer vision applications, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the challenge of training a unified model for all-in-one image restoration, where task interference arises from varying degradation types, by proposing DFPIR with degradation-aware feature perturbations to align feature and parameter spaces, achieving state-of-the-art performance across tasks like denoising, dehazing, deraining, motion deblurring, and low-light enhancement.
All-in-one image restoration aims to recover clear images from various degradation types and levels with a unified model. Nonetheless, the significant variations among degradation types present challenges for training a universal model, often resulting in task interference, where the gradient update directions of different tasks may diverge due to shared parameters. To address this issue, motivated by the routing strategy, we propose DFPIR, a novel all-in-one image restorer that introduces Degradation-aware Feature Perturbations(DFP) to adjust the feature space to align with the unified parameter space. In this paper, the feature perturbations primarily include channel-wise perturbations and attention-wise perturbations. Specifically, channel-wise perturbations are implemented by shuffling the channels in high-dimensional space guided by degradation types, while attention-wise perturbations are achieved through selective masking in the attention space. To achieve these goals, we propose a Degradation-Guided Perturbation Block (DGPB) to implement these two functions, positioned between the encoding and decoding stages of the encoder-decoder architecture. Extensive experimental results demonstrate that DFPIR achieves state-of-the-art performance on several all-in-one image restoration tasks including image denoising, image dehazing, image deraining, motion deblurring, and low-light image enhancement. Our codes are available at https://github.com/TxpHome/DFPIR.