Benchmarking Single Image Dehazing and Beyond

2017/12/12 by Boyi Li, Li, Boyi, Wenqi Ren +11 · 96 citations
Computer Science · #Artificial Intelligence (cs.AI) #Artificial intelligence #Benchmark (surveying) #Benchmarking #Cartography #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Computer vision #Evaluation methods #FOS: Computer and information sciences #Geography #Image (mathematics) #Image Enhancement Techniques #Machine Learning (cs.LG) #Ranging #Scale (ratio) #Task (project management) #Telecommunications #Variety (cybernetics) #Video Surveillance and Tracking Methods #Visual Attention and Saliency Detection #cs.AI #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.1712.04143

published in arXiv (Cornell University) (Cornell University) · IEEE Transactions on Image Processing(TIP 2019)

openalex publication_date 2017/12/12 · arxiv created 2019/04/22 · arxiv updated 2019/04/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06

Abstract

We present a comprehensive study and evaluation of existing single image dehazing algorithms, using a new large-scale benchmark consisting of both synthetic and real-world hazy images, called REalistic Single Image DEhazing (RESIDE). RESIDE highlights diverse data sources and image contents, and is divided into five subsets, each serving different training or evaluation purposes. We further provide a rich variety of criteria for dehazing algorithm evaluation, ranging from full-reference metrics, to no-reference metrics, to subjective evaluation and the novel task-driven evaluation. Experiments on RESIDE shed light on the comparisons and limitations of state-of-the-art dehazing algorithms, and suggest promising future directions.

Citations

• No-Reference Image Quality Assessment in the Spatial Domain
• ImageNet Large Scale Visual Recognition Challenge
• A Statistical Evaluation of Recent Full Reference Image Quality Assessment Algorithms
• Blind Image Quality Assessment: A Natural Scene Statistics Approach in the DCT Domain
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal\n Networks
• Semantic Foggy Scene Understanding with Synthetic Data
• Sparse-to-Dense: Depth Prediction from Sparse Depth Samples and a Single Image
• Rank Analysis of Incomplete Block Designs: I. The Method of Paired Comparisons

Cited by

• Restore-R1: Efficient Image Restoration Agents via Reinforcement Learning with Multimodal LLM Perceptual Feedback
• ClusIR: Towards Cluster-Guided All-in-One Image Restoration
• Unleashing Degradation-Carrying Features in Symmetric U-Net: Simpler and Stronger Baselines for All-in-One Image Restoration
• SFP: Real-World Scene Recovery Using Spatial and Frequency Priors
• Fourier-RWKV: A Multi-State Perception Network for Efficient Image Dehazing
• EvoIR: Towards All-in-One Image Restoration via Evolutionary Frequency Modulation
• Rethinking Surgical Smoke: A Smoke-Type-Aware Laparoscopic Video Desmoking Method and Dataset
• IRPO: Boosting Image Restoration via Post-training GRPO
• Estimating Fog Parameters from a Sequence of Stereo Images
• Test-Time Preference Optimization for Image Restoration
• VLM-Augmented Degradation Modeling for Image Restoration Under Adverse Weather Conditions
• UniSER: A Foundation Model for Unified Soft Effects Removal
• FAPE-IR: Frequency-Aware Planning and Execution Framework for All-in-One Image Restoration
• Learning Implicit Neural Degradation Representation for Unpaired Image Dehazing
• Uni-MoE-2.0-Omni: Scaling Language-Centric Omnimodal Large Model with Advanced MoE, Training and Data
• From Events to Clarity: The Event-Guided Diffusion Framework for Dehazing
• Physically Interpretable Multi-Degradation Image Restoration via Deep Unfolding and Explainable Convolution
• Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start
• Learning to Restore Multi-Degraded Images via Ingredient Decoupling and Task-Aware Path Adaptation
• Delving into Cascaded Instability: A Lipschitz Continuity View on Image Restoration and Object Detection Synergy
• Residual Diffusion Bridge Model for Image Restoration
• Generative AI in Depth: A Survey of Recent Advances, Model Variants, and Real-World Applications
• WaMaIR: Image Restoration via Multiscale Wavelet Convolutions and Mamba-based Channel Modeling with Texture Enhancement
• Pruning Overparameterized Multi-Task Networks for Degraded Web Image Restoration
• Universal Image Restoration Pre-training via Masked Degradation Classification
• Clear Roads, Clear Vision: Advancements in Multi-Weather Restoration for Smart Transportation
• VAMamba: An Efficient Visual Adaptive Mamba for Image Restoration
• WeatherCycle: Unpaired Multi-Weather Restoration via Color Space Decoupled Cycle Learning
• DeLiVR: Differential Spatiotemporal Lie Bias for Efficient Video Deraining
• Unleashing the Potential of the Semantic Latent Space in Diffusion Models for Image Dehazing
• CoRE-UIR: Prior-guided common and residual experts for efficient all-in-one remote sensing image restoration
• What to Remove, What to Preserve: Dual-Ambiguity Rectification for All-in-One Image Restoration
• Degradation-Aware All-in-One Image Restoration via Latent Prior Encoding
• HazeFlow: Revisit Haze Physical Model as ODE and Non-Homogeneous Haze Generation for Real-World Dehazing
• RAM++: Robust Representation Learning via Adaptive Mask for All-in-One Image Restoration
• WeatherBench: A Real-World Benchmark Dataset for All-in-One Adverse Weather Image Restoration
• A Synthetic-to-Real Dehazing Method based on Domain Unification
• Semi-supervised Image Dehazing via Expectation-Maximization and Bidirectional Brownian Bridge Diffusion Models
• Residual-based Efficient Bidirectional Diffusion Model for Image Dehazing and Haze Generation
• TAP: Parameter-efficient Task-Aware Prompting for Adverse Weather Removal
• Diffusion Once and Done: Degradation-Aware LoRA for Efficient All-in-One Image Restoration
• ODOV: Benchmark the Open-Domain Open-Vocabulary Object Detection
• Can Large Pretrained Depth Estimation Models Help With Image Dehazing?
• Towards Robust Semantic Correspondence: A Benchmark and Insights
• UniLDiff: Unlocking the Power of Diffusion Priors for All-in-One Image Restoration
• Event-Based De-Snowing for Autonomous Driving
• Continual Learning-Based Unified Model for Unpaired Image Restoration Tasks
• PHATNet: A Physics-guided Haze Transfer Network for Domain-adaptive Real-world Image Dehazing
• Exploring Scalable Unified Modeling for General Low-Level Vision
• Efficient Dual-domain Image Dehazing with Haze Prior Perception
• When Schrödinger Bridge Meets Real-World Image Dehazing with Unpaired Training
• Hierarchical Semantic-Visual Fusion of Visible and Near-infrared Images for Long-range Haze Removal
• Frequency Domain-Based Diffusion Model for Unpaired Image Dehazing
• Oneta: Multi-Style Image Enhancement Using Eigentransformation Functions
• EAMamba: Efficient All-Around Vision State Space Model for Image Restoration
• Elucidating and Endowing the Diffusion Training Paradigm for General Image Restoration
• Enhancing Image Restoration Transformer via Adaptive Translation Equivariance
• Visual-Instructed Degradation Diffusion for All-in-One Image Restoration
• ADAM-Dehaze: Adaptive Density-Aware Multi-Stage Dehazing for Improved Object Detection in Foggy Conditions
• Zero-Shot Solving of Imaging Inverse Problems via Noise-Refined Likelihood Guided Diffusion Models
• Learning Unpaired Image Dehazing with Physics-based Rehazy Generation
• UniDet-D: A Unified Dynamic Spectral Attention Model for Object Detection under Adverse Weathers
• Image Restoration via Multi-domain Learning
• WDMamba: When Wavelet Degradation Prior Meets Vision Mamba for Image Dehazing
• The Devil is in the Darkness: Diffusion-Based Nighttime Dehazing Anchored in Brightness Perception
• Boosting All-in-One Image Restoration via Self-Improved Privilege Learning
• IRBridge: Solving Image Restoration Bridge with Pre-trained Generative Diffusion Models
• Neural Restoration of Greening Defects in Historical Autochrome Photographs Based on Purely Synthetic Data
• From Controlled Scenarios to Real-World: Cross-Domain Degradation Pattern Matching for All-in-One Image Restoration
• BaryIR: Learning Multi-Source Unified Representation in Continuous Barycenter Space for Generalizable All-in-One Image Restoration
• Weather-Magician: Reconstruction and Rendering Framework for 4D Weather Synthesis In Real Time
• Benchmarking Endoscopic Surgical Image Restoration and Beyond
• Efficient Degradation-agnostic Image Restoration via Channel-Wise Functional Decomposition and Manifold Regularization
• RestoreVAR: Visual Autoregressive Generation for All-in-One Image Restoration
• MODEM: A Morton-Order Degradation Estimation Mechanism for Adverse Weather Image Recovery
• Clear Nights Ahead: Towards Multi-Weather Nighttime Image Restoration
• Fast and Accurate Image Restoration and Generation with Rank Enhanced Linear Attention
• Degradation-Aware Feature Perturbation for All-in-One Image Restoration
• Gated Context Aggregation Network for Image Dehazing and Deraining
• Improved Techniques for Learning to Dehaze and Beyond: A Collective Study
• HorusEye: Language as Dynamic Attention for Emergency Visual Analysis
• Multiple Linear Regression Haze-removal Model Based on Dark Channel Prior
• Real-world Underwater Enhancement: Challenges, Benchmarks, and Solutions
• PAD-Net: A Perception-Aided Single Image Dehazing Network
• Physics-Guided Image Dehazing Diffusion
• DGSolver: Diffusion Generalist Solver with Universal Posterior Sampling for Image Restoration
• Fine-Tuning Adversarially-Robust Transformers for Single-Image Dehazing
• Cross Paradigm Representation and Alignment Transformer for Image Deraining
• FPGA-Based Neural Network Accelerators for Space Applications: A Survey
• Any Image Restoration via Efficient Spatial-Frequency Degradation Adaptation
• JarvisIR: Elevating Autonomous Driving Perception with Intelligent Image Restoration
• Flow-Map Distillation on Relation Manifolds for Image Restoration
• Pseudo-Label Guided Real-World Image De-weathering: A Learning Framework with Imperfect Supervision
• CoA: Towards Real Image Dehazing via Compression-and-Adaptation
• Lumina-OmniLV: A Unified Multimodal Framework for General Low-Level Vision
• DSwinIR: Rethinking Window-based Attention for Image Restoration

Related

• I-HAZE: a dehazing benchmark with real hazy and haze-free indoor images
• A New Multi-Level Hazy Image and Video Dataset for Benchmark of Dehazing Methods
• O-HAZE: a dehazing benchmark with real hazy and haze-free outdoor images
• A Comprehensive Survey and Taxonomy on Single Image Dehazing Based on Deep Learning
• Improved Techniques for Learning to Dehaze and Beyond: A Collective Study