Continual Learning-Based Unified Model for Unpaired Image Restoration Tasks

TL;DR

This paper introduces a unified continual learning framework for unpaired image restoration across various adverse weather conditions, combining innovative feature selection, catastrophic forgetting mitigation, and domain translation techniques.

Contribution

It presents a novel unified model with selective kernel fusion, EWC, and cycle-contrastive loss for multi-weather image restoration, reducing data dependence and improving performance.

Findings

Significant PSNR and SSIM improvements over state-of-the-art methods.

Effective mitigation of catastrophic forgetting across tasks.

Enhanced perceptual quality in restored images.

Abstract

Restoration of images contaminated by different adverse weather conditions such as fog, snow, and rain is a challenging task due to the varying nature of the weather conditions. Most of the existing methods focus on any one particular weather conditions. However, for applications such as autonomous driving, a unified model is necessary to perform restoration of corrupted images due to different weather conditions. We propose a continual learning approach to propose a unified framework for image restoration. The proposed framework integrates three key innovations: (1) Selective Kernel Fusion layers that dynamically combine global and local features for robust adaptive feature selection; (2) Elastic Weight Consolidation (EWC) to enable continual learning and mitigate catastrophic forgetting across multiple restoration tasks; and (3) a novel Cycle-Contrastive Loss that enhances feature discrimination while preserving semantic consistency during domain translation. Further, we propose an unpaired image restoration approach to reduce the dependance of the proposed approach on the training data. Extensive experiments on standard benchmark datasets for dehazing, desnowing and deraining tasks demonstrate significant improvements in PSNR, SSIM, and perceptual quality over the state-of-the-art.