Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start

TL;DR

This paper introduces a dual-level reinforcement learning framework trained on a high-fidelity weather dataset to improve real-world adverse weather image restoration, achieving state-of-the-art results.

Contribution

It presents a novel dual-level reinforcement learning approach with a physics-driven dataset for effective real-world weather image restoration.

Findings

State-of-the-art performance across diverse weather conditions

Effective adaptation without paired supervision

A new high-fidelity weather dataset HFLS-Weather

Abstract

Adverse weather severely impairs real-world visual perception, while existing vision models trained on synthetic data with fixed parameters struggle to generalize to complex degradations. To address this, we first construct HFLS-Weather, a physics-driven, high-fidelity dataset that simulates diverse weather phenomena, and then design a dual-level reinforcement learning framework initialized with HFLS-Weather for cold-start training. Within this framework, at the local level, weather-specific restoration models are refined through perturbation-driven image quality optimization, enabling reward-based learning without paired supervision; at the global level, a meta-controller dynamically orchestrates model selection and execution order according to scene degradation. This framework enables continuous adaptation to real-world conditions and achieves state-of-the-art performance across a wide range of adverse weather scenarios. Code is available at https://github.com/xxclfy/AgentRL-Real-Weather