Blind Image Quality Assessment: A Natural Scene Statistics Approach in the DCT Domain

By Michele A. Saad, A. Bovik, Christophe Charrier. 2012

1,735 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

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

The Perception-Distortion Tradeoff (2017, methodology): “[26] which is specifically designed for SR quality assessment (see Supplementary for experiments with the NR metrics BRISQUE [29], NIQE [30] and BLIINDS-II [37]).”
Learning a No-Reference Quality Metric for Single-Image Super-Resolution (2016, methodology): “In addition, we compare the original results of BRISQUE, BLIINDS, CORNIA, and CNNIQA in Table 10 and Figure 13.”
Fully Deep Blind Image Quality Predictor (2017, methodology): “Based on this observation, percentile pooling strategies have been adopted in IQA studies [12], [31].”
MetaIQA: Deep Meta-Learning for No-Reference Image Quality Assessment (2020, methodology): “Generally, conventional hand-crafted feature-based general-purpose NR-IQA methods can be divided into natural scene statistics (NSS) based metrics [8, 32, 33, 39] and learning-based metrics [53, 54, 36].”
The 2018 PIRM Challenge on Perceptual Image Super-Resolution (2018, methodology): “Note that we also tested the full-reference measures VIF [32], FSIM [49] and MS-SSIM [46], and the no-reference measures CORNIA [48] and BLIINDS [30], which all failed to correctly assess the perceptual quality 4 .”
Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment (2016, methodology): “Following the training and testing protocol outlined in Section III-E, the results achieved by the proposed method on LIVE and TID2013 as summarized in comparison to state-of-the-art general purpose NR IQMs (DIIVINE [13], BLINDS-II [14], CORNIA [16], QAF [18], CNN [19] and SOM [17]) in Table I.”
From QoS to QoE: A Tutorial on Video Quality Assessment (2015, background): “Though some of the objective quality models can realize real-time QoE prediction(e.g., [14]–[27]), it is still an indirect way for QoE prediction.”
Exploring CLIP for Assessing the Look and Feel of Images (2022, methodology): “Conventional NR-IQA methods [29, 31, 32, 45, 66] mainly rely on natural scene statistics (NSS) [44], a model revealing statistic property of natural images.”
Blind Image Quality Estimation via Distortion Aggravation (2018, methodology): “Ten state-of-the-art blind IQA measures act as competitors in this paper, including: (1) NSS based measures, i.e., DIIVINE [15], BLIINDS2 [16], BRISQUE [17], NIQE [28], and ILNIQE [19]; (2) learning based measures, i.e., CORNIA [29] and HOSA [30]; (3) human vision based and some other measures, i.e.”
KADID-10k: A Large-scale Artificially Distorted IQA Database (2019, neutral): “I QA BIQI [24] 0.460 0.431 0.299 BLIINDS-II [25] 0.559 0.527 0.375 BRISQUE [26] 0.554 0.519 0.368 CORNIA * [27] 0.580 0.541 0.384 DIIVINE [28] 0.532 0.489 0.341 HOSA * [29] 0.653 0.609 0.438 SSEQ [30] 0.463 0.424 0.295 InceptionResNetV2 (ﬁne-tune) 0.734 0.731 0.546 * To reduce time complexity,…”
Blind Prediction of Natural Video Quality (2014, background): “…this new blind VQA model are that it 1) characterizes the type of motion in the video, 2) models temporal as well as spatial video attributes, 3) is based on a model of natural video statistics, 4) is computationally fast, and 5) extracts a small number of interpretable features relevant to…”
I-HAZE: a dehazing benchmark with real hazy and haze-free indoor images (2018, methodology): “Other non-reference image quality assessment (NR-IQA) strategies [27], [28], [29] have been used for dehazing assessment.”
Perceptual quality prediction on authentically distorted images using a bag of features approach (2016, methodology): “Existing research on quality assessment has demonstrated that incorporating multiscale information when assessing quality produces quality assessment algorithms that perform better in terms of correlation with human perception (Saad et al., 2012; Wang, Simoncelli, & Bovik, 2003).”
Blind Image Quality Assessment Based on High Order Statistics Aggregation (2016, background): “In [17], Saad et al. train a probabilistic model with statistical features extracted from the DCT domain, but the block processing based algorithm is extraordinarily time consuming.”
No-Reference Image Quality Assessment in the Spatial Domain (2012, methodology): “Our work on NSS based FR QA algorithms [9], [23], [24], more recent RR models [3] and very recent work on NSS based NR QA [12], [13], [25] have led us to the conclusion that visual features derived from NSS lead to particularly potent and simple QA models [26].”
Blind Image Quality Assessment via Deep Learning (2015, methodology): “(c) BLIINDS-II.”