Gradient Magnitude Similarity Deviation: A Highly Efficient Perceptual Image Quality Index

Abstract

It is an important task to faithfully evaluate the perceptual quality of output images in many applications such as image compression, image restoration and multimedia streaming. A good image quality assessment (IQA) model should not only deliver high quality prediction accuracy but also be computationally efficient. The efficiency of IQA metrics is becoming particularly important due to the increasing proliferation of high-volume visual data in high-speed networks. We present a new effective and efficient IQA model, called gradient magnitude similarity deviation (GMSD). The image gradients are sensitive to image distortions, while different local structures in a distorted image suffer different degrees of degradations. This motivates us to explore the use of global variation of gradient based local quality map for overall image quality prediction. We find that the pixel-wise gradient magnitude similarity (GMS) between the reference and distorted images combined with a novel pooling strategy the standard deviation of the GMS map can predict accurately perceptual image quality. The resulting GMSD algorithm is much faster than most state-of-the-art IQA methods, and delivers highly competitive prediction accuracy.

Cited by in corpus (103)

• Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment
• Image Quality Assessment: Unifying Structure and Texture Similarity
• Reduced Reference Perceptual Quality Model and Application to Rate Control for 3D Point Cloud Compression
• dipIQ: Blind Image Quality Assessment by Learning-to-Rank Discriminable Image Pairs
• UGC-VQA: Benchmarking Blind Video Quality Assessment for User Generated Content
• Loss Functions for Neural Networks for Image Processing
• Comparison of Image Quality Models for Optimization of Image Processing Systems
• RAPIQUE: Rapid and Accurate Video Quality Prediction of User Generated Content
• Mean Deviation Similarity Index: Efficient and Reliable Full-Reference Image Quality Evaluator
• FastDeRain: A Novel Video Rain Streak Removal Method Using Directional Gradient Priors
• Perceptual Quality Assessment of Omnidirectional Images
• Towards a quality metric for dense light fields
• Tensor Oriented No-Reference Light Field Image Quality Assessment
• No-Reference Light Field Image Quality Assessment Based on Spatial-Angular Measurement
• Adaptive deep learning framework for robust unsupervised underwater image enhancement
• Subjective Quality Database and Objective Study of Compressed Point Clouds With 6DoF Head-Mounted Display
• Rain Streak Removal for Single Image via Kernel Guided CNN
• Confusing Image Quality Assessment: Towards Better Augmented Reality Experience
• Study of 3D Virtual Reality Picture Quality
• Image Quality Assessment for Magnetic Resonance Imaging
• Efficient No-Reference Quality Assessment and Classification Model for Contrast Distorted Images
• Hitchhiker's Guide to Super-Resolution: Introduction and Recent Advances
• Encoding Visual Sensitivity by MaxPol Convolution Filters for Image Sharpness Assessment
• You Only Train Once: Learning a General Anomaly Enhancement Network with Random Masks for Hyperspectral Anomaly Detection
• BASICS: Broad quality Assessment of Static point clouds In Compression Scenarios
• Predicting the Quality of Compressed Videos with Pre-Existing Distortions
• Locally Adaptive Structure and Texture Similarity for Image Quality Assessment
• SMAPGAN: Generative Adversarial Network Based Semi-Supervised Styled Map Tiles Generating Method
• ChipQA: No-Reference Video Quality Prediction via Space-Time Chips
• Light Field Image Quality Assessment With Auxiliary Learning Based on Depthwise and Anglewise Separable Convolutions
• DeepWSD: Projecting Degradations in Perceptual Space to Wasserstein Distance in Deep Feature Space
• DeepFL-IQA: Weak Supervision for Deep IQA Feature Learning
• Data Readiness for AI: A 360-Degree Survey
• NeRF-NQA: No-Reference Quality Assessment for Scenes Generated by NeRF and Neural View Synthesis Methods
• Quality Evaluation of Arbitrary Style Transfer: Subjective Study and Objective Metric
• Quality Prediction on Deep Generative Images
• Learning to Predict Streaming Video QoE: Distortions, Rebuffering and Memory
• A Detail Based Method for Linear Full Reference Image Quality Prediction
• Gradient-based Feature Extraction From Raw Bayer Pattern Images
• Image inpainting using frequency domain priors
• The Bjøntegaard Bible -- Why your Way of Comparing Video Codecs May Be Wrong
• FAIR: Frequency-aware Image Restoration for Industrial Visual Anomaly Detection
• Reconstruction from edge image combined with color and gradient difference for industrial surface anomaly detection
• From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality
• Zero-shot super-resolution with a physically-motivated downsampling kernel for endomicroscopy
• Light Field Synthesis by Training Deep Network in the Refocused Image Domain
• Center Emphasized Visual Saliency and a Contrast-based Full Reference Image Quality Index
• MMDRFuse: Distilled Mini-Model with Dynamic Refresh for Multi-Modality Image Fusion
• Total Variation with Overlapping Group Sparsity and Lp Quasinorm for Infrared Image Deblurring under Salt-and-Pepper Noise
• MRI super-resolution reconstruction using efficient diffusion probabilistic model with residual shifting
• Deep Optimization model for Screen Content Image Quality Assessment using Neural Networks
• Blind Predicting Similar Quality Map for Image Quality Assessment
• Ambiguity of Objective Image Quality Metrics: A New Methodology for Performance Evaluation
• BVI-VFI: A Video Quality Database for Video Frame Interpolation
• Face Sketch Synthesis Style Similarity:A New Structure Co-occurrence Texture Measure
• Training-Free Synthesized Face Sketch Recognition Using Image Quality Assessment Metrics
• No-Reference Color Image Quality Assessment: From Entropy to Perceptual Quality
• A combined full-reference image quality assessment approach based on convolutional activation maps
• Contrast Enhancement of Brightness-Distorted Images by Improved Adaptive Gamma Correction
• Spoofing Detection on Hand Images Using Quality Assessment
• A Color Intensity Invariant Low Level Feature Optimization Framework for Image Quality Assessment
• Multi-feature 360 Video Quality Estimation
• Deep Neural Networks for Blind Image Quality Assessment: Addressing the Data Challenge
• An Augmented Autoregressive Approach to HTTP Video Stream Quality Prediction
• A new method for structural diagnostics with muon tomography and deep learning
• Subjective Annotation for a Frame Interpolation Benchmark using Artefact Amplification
• A study on the adequacy of common IQA measures for medical images
• Perceptual Image Quality Assessment with Transformers
• A Novel Just-Noticeable-Difference-based Saliency-Channel Attention Residual Network for Full-Reference Image Quality Predictions
• Image Quality Assessment Techniques Show Improved Training and Evaluation of Autoencoder Generative Adversarial Networks
• Deviation Based Pooling Strategies For Full Reference Image Quality Assessment
• Parameter choices in HaarPSI for IQA with medical images
• Wavelength Dependence of Image Quality Metrics and Seeing Parameters and their Relation to Adaptive Optics Performance
• Image Quality Assessment for Omnidirectional Cross-reference Stitching
• Full Reference Screen Content Image Quality Assessment by Fusing Multi-level Structure Similarity
• ER-IQA: Boosting Perceptual Quality Assessment Using External Reference Images
• Learning from Synthetic Data for Opinion-free Blind Image Quality Assessment in the Wild
• PieAPP: Perceptual Image-Error Assessment through Pairwise Preference
• A Note on Efficiency of Downsampling and Color Transformation in Image Quality Assessment
• A comprehensive evaluation of full-reference image quality assessment algorithms on KADID-10k
• On Nearest Neighbors in Non Local Means Denoising
• Learning Transformer Features for Image Quality Assessment
• VTAMIQ: Transformers for Attention Modulated Image Quality Assessment
• Contrast and visual saliency similarity-induced index for assessing image quality
• Quantifying Visual Image Quality: A Bayesian View
• Unsupervised Multi-Clustering and Decision-Making Strategies for 4D-STEM Orientation Mapping
• Region-Adaptive Deformable Network for Image Quality Assessment
• CD2 : Combined Distances of Contrast Distributions for the Assessment of Perceptual Quality of Image Processing
• Boosting Image Quality Assessment Performance: Unsupervised Score Fusion by Deep Maximum a Posteriori Estimation
• Image Super-Resolution Quality Assessment: Structural Fidelity Versus Statistical Naturalness
• Saliency detection based on structural dissimilarity induced by image quality assessment model
• Freqformer: Frequency-Domain Transformer for 3-D Reconstruction and Quantification of Human Retinal Vasculature
• Computer Vision and Metrics Learning for Hypothesis Testing: An Application of Q-Q Plot for Normality Test
• HypervolGAN: An efficient approach for GAN with multi-objective training function
• Blind Quality Assessment for Image Superresolution Using Deep Two-Stream Convolutional Networks
• High-Frequency aware Perceptual Image Enhancement
• Salt and pepper noise removal method based on stationary Framelet transform with non-convex sparsity regularization
• A Shift-insensitive Full Reference Image Quality Assessment Model Based on Quadratic Sum of Gradient Magnitude and LOG signals
• Improving precision of objective image/video quality metrics
• Scoot: A Perceptual Metric for Facial Sketches
• Empirical evaluation of full-reference image quality metrics on MDID database
• Quality-aware Unpaired Image-to-Image Translation
• An Efficient Quality Metric for Video Frame Interpolation Based on Motion-Field Divergence