MetaWeather: Few-Shot Weather-Degraded Image Restoration

TL;DR

MetaWeather is a universal meta-learning approach for weather-degraded image restoration that effectively adapts to unseen weather conditions using few-shot learning and a novel spatial-channel matching algorithm.

Contribution

It introduces a unified meta-learning framework capable of restoring images under diverse and unseen weather conditions with minimal data.

Findings

Outperforms state-of-the-art methods on multiple datasets

Successfully adapts to unseen weather conditions

Demonstrates robustness in real-world scenarios

Abstract

Real-world weather conditions are intricate and often occur concurrently. However, most existing restoration approaches are limited in their applicability to specific weather conditions in training data and struggle to generalize to unseen weather types, including real-world weather conditions. To address this issue, we introduce MetaWeather, a universal approach that can handle diverse and novel weather conditions with a single unified model. Extending a powerful meta-learning framework, MetaWeather formulates the task of weather-degraded image restoration as a few-shot adaptation problem that predicts the degradation pattern of a query image, and learns to adapt to unseen weather conditions through a novel spatial-channel matching algorithm. Experimental results on the BID Task II.A, SPA-Data, and RealSnow datasets demonstrate that the proposed method can adapt to unseen weather conditions, significantly outperforming the state-of-the-art multi-weather image restoration methods.