A Statistical Evaluation of Recent Full Reference Image Quality Assessment Algorithms

By Hamid R. Sheikh, Muhammad Farooq Sabir, A. Bovik. 2006

3,113 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

In Evaluation & Benchmarks

Citations in the corpus

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric (2018, background): “LIVE [50] 29 images 5 traditional continuous .”
Learning a No-Reference Quality Metric for Single-Image Super-Resolution (2016, background): “apply intermediate steps to quantify specific image distortions in [34] rather than SR.”
RankIQA: Learning from Rankings for No-Reference Image Quality Assessment (2017, background): “In learning-based methods, local features are extracted and mapped to the MOS using, for example, Support Machine Regression or Neural Networks [4].”
A Completely Blind Video Integrity Oracle (2016, methodology): “The VIIDEO scores were passed through a logistic non-linearity [44] to map to DMOS before computing LCC.”
Deep Learning of Human Visual Sensitivity in Image Quality Assessment Framework (2017, methodology): “To evaluate the generalization of DeepQA models, they were trained using the LIVE IQA database, and tested on the TID2008 database.”
PIPAL: a Large-Scale Image Quality Assessment Dataset for Perceptual Image Restoration (2020, methodology): “In order to evaluate and develop IQA methods, many datasets have been proposed, such as LIVE [42], CSIQ [27], TID2008 and TID2013 [36,35].”
Blind Image Quality Assessment Based on High Order Statistics Aggregation (2016, methodology): “Except for the commonly-used LIVE database [47], TID2013 database [48], CSIQ database [42], Toyama database [49] and LIVE multiply distorted database (MLIVE) [50], we also tested on one new screen image quality assessment database (SIQAD) [51], to show the generalization ability of the proposed meth”
NTIRE 2017 Challenge on Single Image Super-Resolution: Dataset and Study (2017, background): “We mention also LIVE1 [41, 53], L20 [49, 27], ImageNet [38], Kodak 7 or SuperTex136 [8] that are less popular for single-image SR.”
MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessment (2022, methodology): “Cross-dataset Evaluation on LIVE and TID2013.”
Blind Image Quality Assessment Using a General Regression Neural Network (2011, methodology): “The database reflects the result of a large human study [ 1 ], and includes differential mean opinion scores (DMOS) for each distorted image.”
Perceptual Quality Assessment of Smartphone Photography (2020, methodology): “hk) sive, subjective IQA offers the most reliable way of evaluating image quality through psychophysical experiments [30].”
Hypercomplex Quality Assessment of Multi/Hyperspectral Images (2009, background): “be profitably used to assess the quality of distorted images [4].”
Shift-Net: Image Inpainting via Deep Feature Rearrangement (2018, background): “Actually, both PSNR and SSIM are correlated poorly with human assessment of visual quality [12, 14, 24, 36, 42].”
Unsupervised feature learning framework for no-reference image quality assessment (2012, methodology): “(1) LIVE IQA database: LIVE IQA database [20, 21] has been widely used for evaluating the performance of IQA systems.”
No-Reference Image Quality Assessment in the Spatial Domain (2012, methodology): “(a) 2-D scatter plot between shape and scale parameters obtained by fitting GGD to the empirical distributions of MSCN coefficients of pristine images of Berkeley image segmentation database [33] and simulated distorted images, where similar kinds of distortions as those present in the LIVE IQA data”