Perceptual quality prediction on authentically distorted images using a bag of features approach

By Deepti Ghadiyaram, A. Bovik. Journal of Vision, 2016

331 citations recorded by the source corpus.

Current top-performing blind perceptual image quality prediction models are generally trained on legacy databases of human quality opinion scores on synthetically distorted images. Therefore, they learn image features that effectively predict human visual quality judgments of inauthentic and usually isolated (single) distortions. However, real-world images usually contain complex composite mixtures of multiple distortions. We study the perceptually relevant natural scene statistics of such authentically distorted images in different color spaces and transform domains. We propose a “bag of feature maps” approach that avoids assumptions about the type of distortion(s) contained in an image and instead focuses on capturing consistencies—or departures therefrom—of the statistics of real-world images. Using a large database of authentically distorted images, human opinions of them, and bags of features computed on them, we train a regressor to conduct image quality prediction. We demonstrate the competence of the features toward improving automatic perceptual quality prediction by testing a learned algorithm using them on a benchmark legacy database as well as on a newly introduced distortion-realistic resource called the LIVE In the Wild Image Quality Challenge Database. We extensively evaluate the perceptual quality prediction model and algorithm and show that it is able to achieve good-quality prediction power that is better than other leading models.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/336f1e9e3204c478488ac5433c9833378ed66a1a

How researchers see this paper

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

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality (2019, background): “Examples include BRISQUE [10], NIQE [11], CORNIA [13], FRIQUEE [12], which use “handcrafted” statistical features to drive shallow learners (SVM, etc.).”
No-Reference Image Quality Assessment via Transformers, Relative Ranking, and Self-Consistency (2021, neutral): “Before the raise of deep learning, early version of general-purpose NR-IQA methods mainly divide into natu-ral scene statistics (NSS) based metrics [16, 17, 6, 5, 18, 19, 8, 20, 21] and learning-based metrics [22, 23, 24, 25, 26, 27].”
Two-Level Approach for No-Reference Consumer Video Quality Assessment (2019, methodology): “FRIQUEE [43] perform better than V-BLIINDS on consumer”
MetaIQA: Deep Meta-Learning for No-Reference Image Quality Assessment (2020, methodology): “NR-IQA can be classified into distortion-specific methods [25, 24, 29, 47] and general-purpose methods [39, 32, 56, 51, 50, 12, 54, 49, 10].”
Perceptual Quality Assessment of Smartphone Photography (2020, methodology): “These cover a wide range of design philoso-phies, including NSS-based [6, 22, 24, 40], codebook-based [38, 39], and DNN-based [41] models.”
Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wild (2020, background): “Later models aimed for general-purpose BIQA [20]–[24], with the underlying assumption that statistics extracted from natural images are highly regular [25] and distortions will break such statistical regularities.”
MUSIQ: Multi-scale Image Quality Transformer (2021, background): “But since each crop is only method SRCC PLCC DIIVINE [28] 0.599 0.600 BRISQUE [26] 0.809 0.817 CORNIA [42] 0.709 0.725 QAC [40] 0.092 0.497 ILNIQE [47] 0.713 0.721 FRIQUEE [14] 0.819 0.830 DBCNN [49] 0.911 0.915 Fang et al .”
MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessment (2022, methodology): “Previous general-purpose NR-IQA methods are mainly divided into natural scene statistics (NSS) based metrics [7,8,29–31, 34, 46, 55] and learning-based metrics [25, 48–50, 54, 57].”
UGC-VQA: Benchmarking Blind Video Quality Assessment for User Generated Content (2020, methodology): “baseline models include NIQE [53], ILNIQE [78], VIIDEO [55], BRISQUE [38], GM-LOG [57], HIGRADE [40], FRIQUEE [41], CORNIA [43], HOSA [61], KonCept512 [70], PaQ-2-PiQ [64], V-BLIINDS [39], and TLVQM [42].”