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

• URL: http://arxiv.org/abs/2511.05095v1
• Date: Fri, 07 Nov 2025 09:22:53 GMT
• Title: Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start
• Authors: Fuyang Liu, Jiaqi Xu, Xiaowei Hu,
• Abstract summary: Adverse weather severely impairs real-world visual perception, while existing vision models struggle to generalize to complex degradations.<n>We first construct HFLS-Weather, a physics-driven dataset that simulates diverse weather phenomena, and then design a dual-level reinforcement learning framework with HFLS-Weather for cold-start training.<n>This framework enables continuous adaptation to real-world conditions and achieves state-of-the-art performance across a wide range of adverse weather scenarios.
• Score: 12.423595039024597
• License: http://creativecommons.org/licenses/by/4.0/
• 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

Related papers

• WIND: Weather Inverse Diffusion for Zero-Shot Atmospheric Modeling [19.471234487904514]
  We introduce WIND, a single pre-trained foundation model capable of replacing specialized baselines across a vast array of tasks.<n>We frame diverse domain-specific problems strictly as inverse problems and solve them via posterior sampling.<n>We demonstrate the model's capacity to generate physically consistent counterfactual storylines of extreme weather events under global warming scenarios.
  arXiv  Detail & Related papers  (2026-02-03T18:58:10Z)
• WeatherCycle: Unpaired Multi-Weather Restoration via Color Space Decoupled Cycle Learning [30.62082910458533]
  Unsupervised image restoration under multi-weather conditions remains a fundamental yet underexplored challenge.<n>We propose textbfWeatherCycle, a unified framework that reformulates weather restoration as a bidirectional degradation-content translation cycle.<n>Our method achieves state-of-the-art performance among unsupervised approaches, with strong generalization to complex weather degradations.
  arXiv  Detail & Related papers  (2025-09-27T06:44:27Z)
• Pseudo-Label Guided Real-World Image De-weathering: A Learning Framework with Imperfect Supervision [57.5699142476311]
  We propose a unified solution for real-world image de-weathering with non-ideal supervision.<n>Our method exhibits significant advantages when trained on imperfectly aligned de-weathering datasets.
  arXiv  Detail & Related papers  (2025-04-14T07:24:03Z)
• SemiDDM-Weather: A Semi-supervised Learning Framework for All-in-one Adverse Weather Removal [57.52777076116241]
  Adverse weather removal aims to restore clear vision under adverse weather conditions. This paper presents a pioneering semi-supervised all-in-one adverse weather removal framework built on the teacher-student network.
  arXiv  Detail & Related papers  (2024-09-29T12:12:22Z)
• Towards Real-World Adverse Weather Image Restoration: Enhancing Clearness and Semantics with Vision-Language Models [85.53913950137175]
  We formulate a semi-supervised learning framework employing vision-language models to enhance restoration performance in real-world conditions. For clearness enhancement, we use real-world data, utilizing a dual-step strategy with pseudo-labels assessed by vision-language models and weather prompt learning. For semantic enhancement, we integrate real-world data by adjusting weather conditions in vision-language model descriptions while preserving semantic meaning.
  arXiv  Detail & Related papers  (2024-09-03T17:56:51Z)
• On the Integration of Spatial-Temporal Knowledge: A Lightweight Approach to Atmospheric Time Series Forecasting [24.119776558530983]
  The paper emphasizes the effectiveness of spatial-temporal knowledge integration over complex architectures, providing novel insights for atmospheric time series forecasting (ATSF)<n>With 10k parameters and one hour of training, STELLA achieves superior performance on five datasets compared to other advanced methods.
  arXiv  Detail & Related papers  (2024-08-19T04:23:40Z)
• Learning Robust Precipitation Forecaster by Temporal Frame Interpolation [65.5045412005064]
  We develop a robust precipitation forecasting model that demonstrates resilience against spatial-temporal discrepancies. Our approach has led to significant improvements in forecasting precision, culminating in our model securing textit1st place in the transfer learning leaderboard of the textitWeather4cast'23 competition.
  arXiv  Detail & Related papers  (2023-11-30T08:22:08Z)
• WeatherDepth: Curriculum Contrastive Learning for Self-Supervised Depth Estimation under Adverse Weather Conditions [42.99525455786019]
  We propose WeatherDepth, a self-supervised robust depth estimation model with curriculum contrastive learning. The proposed solution is proven to be easily incorporated into various architectures and demonstrates state-of-the-art (SoTA) performance on both synthetic and real weather datasets.
  arXiv  Detail & Related papers  (2023-10-09T09:26:27Z)
• MetaWeather: Few-Shot Weather-Degraded Image Restoration [17.63266150036311]
  We introduce MetaWeather, a universal approach that can handle diverse and novel weather conditions with a single unified model. We show that MetaWeather can adapt to unseen weather conditions, significantly outperforming state-of-the-art multi-weather image restoration methods.
  arXiv  Detail & Related papers  (2023-08-28T06:25:40Z)
• Physics-Driven Turbulence Image Restoration with Stochastic Refinement [80.79900297089176]
  Image distortion by atmospheric turbulence is a critical problem in long-range optical imaging systems. Fast and physics-grounded simulation tools have been introduced to help the deep-learning models adapt to real-world turbulence conditions. This paper proposes the Physics-integrated Restoration Network (PiRN) to help the network to disentangle theity from the degradation and the underlying image.
  arXiv  Detail & Related papers  (2023-07-20T05:49:21Z)
• ClimaX: A foundation model for weather and climate [51.208269971019504]
  ClimaX is a deep learning model for weather and climate science. It can be pre-trained with a self-supervised learning objective on climate datasets. It can be fine-tuned to address a breadth of climate and weather tasks.
  arXiv  Detail & Related papers  (2023-01-24T23:19:01Z)
• Competitive Simplicity for Multi-Task Learning for Real-Time Foggy Scene Understanding via Domain Adaptation [17.530091734327296]
  We propose a multi-task learning approach capable of performing in real-time semantic scene understanding and monocular depth estimation under foggy weather conditions. Our model incorporates RGB colour, depth, and luminance images via distinct encoders with dense connectivity and features fusing.
  arXiv  Detail & Related papers  (2020-12-09T20:38:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.