MetaWeather: Few-Shot Weather-Degraded Image Restoration
This work addresses the challenge of generalizing image restoration to real-world, concurrent weather conditions, which is incremental as it builds on meta-learning frameworks.
The authors tackled the problem of image restoration under diverse and unseen weather conditions by introducing MetaWeather, a universal model that adapts to novel weather types using few-shot learning, achieving significant performance improvements over state-of-the-art methods on multiple datasets.
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.