VSI: A Visual Saliency-Induced Index for Perceptual Image Quality Assessment

By Lin Zhang, Ying Shen, Hongyu Li. IEEE Transactions on Image Processing, 2014

980 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/d15c847414e71cdfc945cbd27db53d6b193e093c

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

Citations in the corpus

An Experiment-Based Review of Low-Light Image Enhancement Methods (2020, methodology): “…[303], the Gradient Magnitude Similarity Deviation (GMSD) [304], the visual information ﬁdelity (VIF) [305], the visual saliency index (VSI) [306], the tone-mapped image quality index (TMQI) [307], the Statistical Naturalness Measure (SNM) [307], and the Feature SIMilarity Index (FSIM)…”
No-Reference Quality Metric of Contrast-Distorted Images Based on Information Maximization (2017, neutral): “FR-IQA metrics involve state-ofthe-art FSIM [2], IGM [5], LTG [7] and VSI [8].”
PieAPP: Perceptual Image-Error Assessment Through Pairwise Preference (2018, methodology): “As is common in other IQA algorithms [7,56,57], we compute errors on a patchwise basis by feeding corresponding patches from A, B, and R through the FE and SC subnets, and aggregate them to obtain the overall errors, sA and sB.”
From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality (2019, background): “Full-Reference models: Many full-reference (FR) perceptual picture quality predictors, which make comparisons against high-quality reference pictures, are available [9, 10], [27, 28, 29, 30, 31, 32, 33].”
Blind Quality Assessment Based on Pseudo-Reference Image (2018, methodology): “We have tested several representative FR metrics including: SSIM [53], MS-SSIM [54], VIF [55], FSIM [56], VSI [57], MDSI [58], PSIM [59] and GMSD [49].”
Using labeled data to evaluate change detectors in a multivariate streaming environment (2009, positive): “EDANet [17] also utilizes the strategy of dense connection, and the difference is that EDANet adopts the dense connection between modules using asymmetric convolution.”
Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment (2016, neutral): “The Visual Saliency-Induced Index (VSI) [32] takes the local saliency from reference and distorted image as features maps in a similarity formalism (Section II-A)”
A Haar wavelet-based perceptual similarity index for image quality assessment (2016, methodology): “Furthermore, [13] also explores the influence of different saliency models on the performance of the proposed image quality measure.”
Domain Fingerprints for No-Reference Image Quality Assessment (2019, neutral): “BOTH PLCC AND SROCC ON THE LIVE DATASET. Category Method TID2013 LIVE SROCC PLCC SROCC PLCC FR-IQA PSNR 0.889 0.847 0.880 0.805 SSIM [30] 0.856 0.867 0.918 0.780 FSIM [31] 0.963 0.932 0.952 0.822 VSI [32] 0.947 0.939 0.936 0.853 NR-IQA DIVINE [8] 0.855 0.851 0.885 0.853 BLIINDS-II [33] 0.877 0.841 0”
A Patch-Structure Representation Method for Quality Assessment of Contrast Changed Images (2015, methodology): “The PCQI map produced by the proposed method also indicates local quality variations, which are useful in guiding the future design of advanced image enhancement algorithms.”
Perceptual Quality Assessment for Multi-Exposure Image Fusion (2015, background): “These include information theoretic and natural scene statistical approaches [38]–[40], the adaptive linear system decomposition framework [41]–[43], the feature similarity method [44], and visual attention and saliency-based approaches [39], [45]–[47].”
KADID-10k: A Large-scale Artificially Distorted IQA Database (2019, neutral): “By changing [14] 0.801 0.802 0.609 IWSSIM [15] 0.846 0.850 0.666 MDSI [16] 0.873 0.872 0.682 VSI [17] 0.878 0.879 0.691 FSIM [18] 0.851 0.854 0.665 GMSD [19] 0.847 0.847 0.664 SFF [20] 0.862 0.862 0.675 SCQI [21] 0.853 0.854 0.662 ADD-GSIM [22] 0.817 0.818 0.621 SR-SIM [23] 0.834 0.839 0.652 N R -”
Exploring CLIP for Assessing the Look and Feel of Images (2022, background): “The main limitation of FR-IQA is that it is conﬁned to scenarios where references exist.”
Learning a No-Reference Quality Assessment Model of Enhanced Images With Big Data (2018, methodology): “In this comparison, apart from our BIQME method, we mainly consider the following 13 state-of-the-art IQA models, which encompass: 1) FR FSIM [8], LTG [9], VSI [10], PSIM [11], and C-PCQI; 2) RR RRED [15], FTQM [16], RIQMC [30], and QMC [34]; and 3) NR NIQE [19], IL-NIQE [20],”
Fully Deep Blind Image Quality Predictor (2017, methodology): “Here, four FR-IQA metrics are adopted to generate local quality maps: structural similarity index (SSIM) [26], gradient magnitude similarity deviation (GMSD) [27], feature similarity index (FSIM) [51], and visual saliency-based index (VSI) [52].”
Image Quality Assessment: Unifying Structure and Texture Similarity (2020, neutral): “605 VSI [64] 0.”
The Analysis of Image Contrast: From Quality Assessment to Automatic Enhancement (2016, methodology): “mechanism (IGM) [13], spectral residual (SR) SR-SSIM [14], and VS-based index (VSI) [15] have been introduced”
TOPIQ: A Top-Down Approach From Semantics to Distortions for Image Quality Assessment (2023, positive): “To address this, Wang et al. [13] introduced the structural similarity (SSIM) index to compare structural similarity in local patches, which inspired a lot of follow-up works [7], [14], [15], [16], [17], [18], [19].”