Backbone-Agnostic Stochastic Perturbation Learning for End-to-End Real-World Image Dehazing

• URL: http://arxiv.org/abs/2607.11623v2
• Date: Thu, 16 Jul 2026 03:29:58 GMT
• Title: Backbone-Agnostic Stochastic Perturbation Learning for End-to-End Real-World Image Dehazing
• Abstract summary: We propose Backbone-Agnostic Perturbation Learning, a plug-and-play framework for end-to-end real-world image dehazing.<n> BSPL consistently improves multiple backbone representatives with only marginal additional inference overhead.
• Score: 2.3013099757243056
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Real-world paired image dehazing remains challenging because haze degradation is spatially non-uniform, illumination-dependent, and physically ambiguous even when haze-free references are available. Existing end-to-end restoration networks usually learn a deterministic mapping from a hazy observation to a clean target, while degradation-sensitive feature responses, reverse haze-formation consistency, and cross-domain negative structure remain insufficiently exploited. In this paper, we propose Backbone-Agnostic Stochastic Perturbation Learning (BSPL), a plug-and-play framework for end-to-end real-world image dehazing. BSPL first introduces a Learnable Stochastic Perturbation Modulator (LSPM), which learns input-conditioned channel-wise and spatial-wise perturbation distributions and converts the resulting feature-response discrepancies into adaptive modulation weights. It then develops a Prior-informed Perturbation-guided Reconstruction Module (PPRM), which reuses the learned bottleneck perturbations together with transmission and atmospheric-light priors to reconstruct the hazy observation from the restored result and enforce degradation consistency. Furthermore, we propose a Dual-space Domain-diversified Distribution-aware Contrastive Loss ($D^3$CL) to regularize both clean restoration and hazy reconstruction spaces with real-world and synthetic negatives. Experiments on five real-world paired benchmarks show that BSPL consistently improves multiple representative backbones with only marginal additional inference overhead.

Related papers

• 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)
• Clipping Bottleneck: Stabilizing RLVR via Stochastic Recovery of Near-Boundary Signals [83.0127582612634]
  Near-boundary Rescue (NSR) is a minimal, plug-and-play modification that retains out-of-bound tokens to recover lost signals.<n>NSR substantially improves training stability and delivers consistent gains over strong baselines such as DAPO and GSPO.
  arXiv  Detail & Related papers  (2026-05-21T16:45:31Z)
• Next-Acceleration-Scale Prediction for Autoregressive MRI Reconstruction [52.32112533846212]
  MRI reconstruction is an inherently ill-posed inverse problem, since incomplete measurements admit many plausible solutions.<n>We address this limitation by moving reconstruction to discrete multi-scale latent space and posing it as autoregressive next-acceleration-scale prediction.<n>We show that our approach delivers improved reconstruction performance across diverse sampling patterns under extreme undersampling.
  arXiv  Detail & Related papers  (2026-05-19T04:40:50Z)
• HIR-ALIGN: Enhancing Hyperspectral Image Restoration via Diffusion-Based Data Generation [50.46910397782266]
  Hyperspectral image (HSI) restoration is crucial for reliable analysis, as real HSIs suffer from degradations like noise, blur, and resolution loss.<n>We present HIR-ALIGN, a plug-and-play target-adaptive augmentation framework that enhances hyperspectral image restoration by augmenting limited training images with synthetic data that closely matches the target distribution using no extra data.
  arXiv  Detail & Related papers  (2026-05-13T14:14:13Z)
• ZeroIDIR: Zero-Reference Illumination Degradation Image Restoration with Perturbed Consistency Diffusion Models [72.4683657127516]
  We propose a zero-reference diffusion-based framework, named ZeroIDIR, for illumination image restoration.<n>The framework decouples the restoration process into adaptive illumination correction and diffusion-based reconstruction.<n>Experiments on publicly available benchmarks show that the proposed method outperforms state-of-the-art unsupervised competitors.
  arXiv  Detail & Related papers  (2026-05-12T02:36:49Z)
• Continuous Expert Assembly: Instance-Conditioned Low-Rank Residuals for All-in-One Image Restoration [41.34371852767098]
  CEA is a token-wise dynamic parameterization framework for all-in-one image restoration.<n>CEA improves average restoration quality over strong prompt-, descriptor-, and expert-based baselines.
  arXiv  Detail & Related papers  (2026-05-07T12:31:24Z)
• Beyond Reconstruction: Reconstruction-to-Vector Diffusion for Hyperspectral Anomaly Detection [9.078979356836193]
  Hyperspectral Anomaly Detection (HAD) excels at identifying sparse targets in complex scenes.<n>Existing models remain trapped in a scalar "reconstruction-as-endpoint" paradigm.<n>We propose Reconstruction-to-Diffusion (R2VD), which redefines reconstruction as a purification origin.<n>R2VD establishes a new state-of-the-art, delivering exceptional target detectability background.
  arXiv  Detail & Related papers  (2026-04-13T12:31:50Z)
• PRISM: Rethinking Scattered Atmosphere Reconstruction as a Unified Understanding and Generation Model for Real-world Dehazing [22.670768613999787]
  PRISM is a physically structured framework that jointly reconstructs the clear scene and scattering variables under the atmospheric scattering model.<n>We design an online non-uniform haze synthesis pipeline and a Selective Self-distillation Adaptation scheme for unpaired real-world scenarios.<n>Experiments on real-world benchmarks demonstrate that PRISM achieves state-of-the-art performance on RID tasks.
  arXiv  Detail & Related papers  (2026-04-08T13:01:30Z)
• Customizing Large Vision Model-Guided Low-Rank Approximation for Ground-Roll Denoise [4.536244262111516]
  Ground-roll is a dominant source of coherent noise in land and vertical seismic profiling (VSP) data.<n>We propose a training-free framework that reformulates ground-roll attenuation as a semantic-guided signal separation problem.
  arXiv  Detail & Related papers  (2026-04-01T14:59:48Z)
• FUMO: Prior-Modulated Diffusion for Single Image Reflection Removal [51.98991734660274]
  This paper presents a diffusion model with prior modulation framework (FUMO) that introduces explicit guidance signals to improve spatial controllability and structural faithfulness.<n>Experiments conducted on both standard benchmarks and challenging images in the wild demonstrate competitive quantitative results and consistently improved perceptual quality.
  arXiv  Detail & Related papers  (2026-03-19T15:33:11Z)
• U-DAVI: Uncertainty-Aware Diffusion-Prior-Based Amortized Variational Inference for Image Reconstruction [10.273906387994902]
  Ill-posed imaging inverse problems remain challenging due to the ambiguity in mapping degraded observations to clean images.<n>Amortized variational inference frameworks address this inefficiency by learning a direct mapping from measurements to posteriors.<n>We extend the amortized framework by injecting spatially adaptive perturbations to measurements during training, guided by uncertainty estimates, to emphasize learning in the most uncertain regions.
  arXiv  Detail & Related papers  (2026-02-12T08:32:11Z)
• AODDiff: Probabilistic Reconstruction of Aerosol Optical Depth via Diffusion-based Bayesian Inference [12.981319326856495]
  AODDiff is a probabilistic reconstruction framework based on diffusion-based inference.<n>It can be flexibly adapted to various reconstruction tasks without requiring task-specific retraining.<n>As a generative model, AODDiff inherently enables uncertainty sampling via multiple spatialquality sampling, offering critical confidence metrics for downstream applications.
  arXiv  Detail & Related papers  (2025-12-31T13:16:10Z)
• Degradation-Aware Metric Prompting for Hyperspectral Image Restoration [20.8230102672152]
  We propose a Degradation-Aware Metric Prompting (DAMP) framework to guide restoration of hyperspectral images.<n>Instead of relying on predefined degradation priors, we design spatial-spectral degradation metrics to continuously quantify degradations.<n>The framework enables adaptive, efficient, and robust restoration under diverse, mixed, or unseen degradations.
  arXiv  Detail & Related papers  (2025-12-23T11:05:36Z)
• Delving into Cascaded Instability: A Lipschitz Continuity View on Image Restoration and Object Detection Synergy [95.93943805282868]
  Lipschitz-regularized object detection (LROD)<n>We propose Lipschitz-regularized YOLO (LR-YOLO), a framework that integrates image restoration directly into the detector's feature learning.<n> experiments on haze and low-light benchmarks demonstrate that LR-YOLO consistently improves detection stability, optimization smoothness, and overall accuracy.
  arXiv  Detail & Related papers  (2025-10-28T09:41:42Z)
• Frequency Domain-Based Diffusion Model for Unpaired Image Dehazing [92.61216319417208]
  We propose a novel frequency domain-based diffusion model, named ours, for fully exploiting the beneficial knowledge in unpaired clear data.<n>Inspired by the strong generative ability shown by Diffusion Models (DMs), we tackle the dehazing task from the perspective of frequency domain reconstruction.
  arXiv  Detail & Related papers  (2025-07-02T01:22:46Z)
• Unsupervised Image Super-Resolution Reconstruction Based on Real-World Degradation Patterns [4.977925450373957]
  We propose a novel TripleGAN framework for training super-resolution reconstruction models.<n>The framework learns real-world degradation patterns from LR observations and synthesizes datasets with corresponding degradation characteristics.<n>Our method exhibits clear advantages in quantitative metrics while maintaining sharp reconstructions without over-smoothing artifacts.
  arXiv  Detail & Related papers  (2025-06-20T14:24:48Z)
• Boosting All-in-One Image Restoration via Self-Improved Privilege Learning [72.35265021054471]
  Self-Improved Privilege Learning (SIPL) is a novel paradigm that overcomes limitations by extending the utility of privileged information (PI) beyond training into the inference stage.<n>Central to SIPL is Proxy Fusion, a lightweight module incorporating a learnable Privileged Dictionary.<n>Extensive experiments demonstrate that SIPL significantly advances the state-of-the-art on diverse all-in-one image restoration benchmarks.
  arXiv  Detail & Related papers  (2025-05-30T04:36:52Z)

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.