Uncertainty-Guided Adverse Weather Restoration via Gated Transformer Network

• URL: http://arxiv.org/abs/2609.02434v1
• Date: Wed, 02 Sep 2026 10:57:41 GMT
• Title: Uncertainty-Guided Adverse Weather Restoration via Gated Transformer Network
• Abstract summary: Restoring images degraded by adverse weather remains challenging due to spatially heterogeneous degradations.<n>We propose an Uncertainty-guided Adverse-weather Restoration Network (UAR-Net) that integrates a gated transformer with balanced multi-scale skip connections.<n>Our method achieves state-of-the-art performance across multiple adverse-weather benchmarks.
• Score: 35.350043473549405
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Restoring images degraded by adverse weather remains challenging due to spatially heterogeneous degradations. Many existing weather-specific restoration models rely on weather-agnostic global aggregation, naive cross-scale fusion, and deterministic objectives, which struggle to handle heterogeneous degradations in all-in-one adverse-weather settings. To address these limitations, we propose an Uncertainty-guided Adverse-weather Restoration Network (UAR-Net), a weather-specific AiO framework that integrates a gated transformer with balanced multi-scale skip connections. Specifically, we employ Gated Dual-scale Transformer Blocks (GDTB) to jointly model selective global interactions and multi-scale local structures, a progressive Balanced Multi-scale Skip Connection (BMSC) for balanced multi-scale feature integration, and an Uncertainty-Aware Refinement Head (URH) that performs artifact removal, detail enhancement, and predictive uncertainty estimation. The model is supervised by a Brightness-Aware Energy Loss (BAE-Loss) to encourage accurate reconstruction with well-calibrated uncertainty. Extensive experiments demonstrate that our method achieves state-of-the-art performance across multiple adverse-weather benchmarks. The codes will open source upon acceptance.

Related papers

• CoRE-UIR: Prior-guided common and residual experts for efficient all-in-one remote sensing image restoration [23.98937821382067]
  CoRE-UIR (Common and Residual Experts for Universal Image Restoration) is a prior-guided global-local framework centered on the Common-and-Residual Expert Block (CoRE)<n>CoRE decomposes restoration capacity into a common dense expert for degradation-invariant restoration and low-rank residual experts for degradation-specific compensation, enabling adaptive specialization without redundant expert replication.<n>Experiments show that CoRE-UIR improves the overall average PSNR by 1.05 dB while running 11.83$times$ faster and reducing peak memory by 85.3% relative to the strongest baseline, BaryIR.
  arXiv  Detail & Related papers  (2026-07-30T09:14:08Z)
• Where to Refine, When to Stop: Rethinking Redundancy via Latent Discrepancy for Efficient Visual Autoregressive Generation [50.139984798361375]
  We propose a training-free framework that removes redundancy via latent discrepancy.<n> LD-Pruning substantially reduces inference while maintaining high generation quality, achieving up to 2.35x speedup on Infinity-8B.
  arXiv  Detail & Related papers  (2026-05-29T19:34:39Z)
• PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow [69.40969296005794]
  We propose a unified framework that integrates zero-shot soft weather perceptions with velocity-constrained rectified-flow refinement.<n>PVRF improves both fidelity and perceptual quality over state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-05-13T19:07:14Z)
• NimbusGS: Unified 3D Scene Reconstruction under Hybrid Weather [83.14198482357388]
  We present NimbusGS, a unified framework for reconstructing high-quality 3D scenes from degraded multi-view inputs.<n>Unlike existing methods that target specific weather types, NimbusGS addresses the broader challenge of generalization.
  arXiv  Detail & Related papers  (2026-03-28T10:46:29Z)
• CAWM-Mamba: A unified model for infrared-visible image fusion and compound adverse weather restoration [8.400835004298624]
  Multimodal Image Fusion (MMIF) integrates complementary information from various modalities to produce clearer and more informative fused images.<n>Existing adverse weather fusion methods only tackle single types of degradation such as haze, rain, or snow, and fail when multiple degradations coexist.<n>We propose Compound Adverse Weather Mamba, the first end-to-end framework that jointly performs image fusion and compound weather restoration with unified shared weights.
  arXiv  Detail & Related papers  (2026-03-03T03:27:05Z)
• CASCADE: Cross-scale Advective Super-resolution with Climate Assimilation and Downscaling Evolution [51.56484100374058]
  CASCADE reconstructs fine structure by iteratively advecting coarse information along learned, flow-conditioned velocity warping fields.<n> evaluated on SEVIR radar data for 4x super-resolution of severe convective storms.<n> produces temporally coherent, physically consistent, and mass-conserving reconstructions well suited to advection-dominated extremes in atmospheric and oceanic applications.
  arXiv  Detail & Related papers  (2026-02-18T14:41:54Z)
• 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)
• Towards Explicit Geometry-Reflectance Collaboration for Generalized LiDAR Segmentation in Adverse Weather [58.4718010073085]
  Existing LiDAR segmentation models often suffer from decreased accuracy when exposed to adverse weather conditions.<n>Recent methods addressing this issue focus on enhancing training data through weather simulation or universal augmentation techniques.<n>We propose a novel Geometry-Reflectance Collaboration framework that explicitly separates feature extraction for geometry and reflectance.
  arXiv  Detail & Related papers  (2025-06-03T03:23:43Z)
• Gradient as Conditions: Rethinking HOG for All-in-one Image Restoration [23.153283910821862]
  Histogram of Oriented Gradients (HOG) as a classical gradient representation has strong discriminative capability across diverse degradations.<n>We propose HOGformer, a Transformer-based model that integrates learnable HOG features for degradation-aware restoration.<n> HOGformer achieves state-of-the-art performance and generalizes well to complex real-world scenarios.
  arXiv  Detail & Related papers  (2025-04-12T23:52:59Z)
• Joint Conditional Diffusion Model for Image Restoration with Mixed Degradations [29.14467633167042]
  We propose a new method for image restoration in adverse weather conditions. We use a mixed degradation model based on atmosphere scattering model to guide the whole restoration process. Experiments on both multi-weather and weather-specific datasets demonstrate the superiority of our method over state-of-the-art competing methods.
  arXiv  Detail & Related papers  (2024-04-11T14:07:16Z)
• Restoring Images Captured in Arbitrary Hybrid Adverse Weather Conditions in One Go [2.0054257354429925]
  We present a novel unified framework, dubbed RAHC, to Restore Arbitrary Hybrid adverse weather Conditions. We also construct a new dataset, HAC, for learning and benchmarking arbitrary Hybrid Adverse Conditions restoration.
  arXiv  Detail & Related papers  (2023-05-17T06:42:42Z)

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.