Saved in:
Bibliographic Details
Main Authors: Liu, Fuyang, Xu, Jiaqi, Hu, Xiaowei
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.05095
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914142463459328
author Liu, Fuyang
Xu, Jiaqi
Hu, Xiaowei
author_facet Liu, Fuyang
Xu, Jiaqi
Hu, Xiaowei
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2511_05095
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start
Liu, Fuyang
Xu, Jiaqi
Hu, Xiaowei
Computer Vision and Pattern Recognition
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
title Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.05095