Making a “Completely Blind” Image Quality Analyzer

By Anish Mittal, Rajiv Soundararajan, A. Bovik. 2013

6,568 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

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution (2021, methodology): “Since the ground-truth for the RealSRSet dataset is not available, we adopt the non-reference image quality assessment (IQA) metrics including NIQE [35], NRQM [29] and PI [4] for quantitative evaluation.”
High-Fidelity Generative Image Compression (2020, methodology): “We find that in the non-conditional setting, SpectralNorm reduces the variability of FID and KID significantly, and, to a lesser extent, also that of LPIPS and NIQE.”
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]).”
PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Models (2020, methodology): “To provide another measure of perceptual quality, we evaluated the Naturalness Image Quality Evaluator (NIQE) score [16], previously used in perceptual super-resolution [10, 4, 24].”
Generalization of the Dark Channel Prior for Single Image Restoration (2018, background): “16) with different color tones and lighting, and use NIQE and two other no-reference quality metrics, Underwater Authorized licensed use limited to: Univ of Calif San Diego.”
2018 PIRM Challenge on Perceptual Image Super-resolution (2018, background): “As above, RMSE, SSIM and IFC are anti -correlated with human-opinion-scores, while NIQE and PI are most correlated (especially in the high perceptual quality regime). altogether), where we average only over diﬀerent human raters.”
Deblurring via Stochastic Refinement (2021, methodology): “We evaluate our method on four different perceptual metrics: LPIPS [77], NIQE [49], FID (Fréchet Inception Distance) [24], and KID (Kernel Inception Distance) [4].”
Blind Prediction of Natural Video Quality (2014, methodology): “The complete list of features used for video quality prediction is: the motion coherency measure and the global motion measure which are key characterizations of the temporal behavior exhibited by a video sequence, the five NVS γ (shape-parameter) spectral ratios, absolute temporal derivative of mea”
MSR-net: Low-light Image Enhancement Using Deep Convolutional Network (2017, methodology): “Since the ground truth is known for the synthetic test data, we use SSIM [31] for a quantitative evaluation and NIQE [22] to assess the natural preservation.”
Quality Aware Generative Adversarial Networks (2019, methodology): “NIQE [5] is a popular NR IQA metric that is based on the statistics of mean subtracted and contrast normalized (MSCN) natural scenes.”
Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data (2021, methodology): “Quantitative Comparisons We provide the non-reference image quality assessment - NIQE [8] for reference.”
EnlightenGAN: Deep Light Enhancement Without Paired Supervision (2019, positive): “t and LIME are not well scored, because of causing many over-exposures and sometimes amplifying the noise. 4.3.3 No-Referenced Image Quality Assessment We adopt Natural Image Quality Evaluator (NIQE) [48], a well-known no-reference image quality assessment for evaluating real image restoration witho”
Frequency Separation for Real-World Super-Resolution (2019, methodology): “We use the pre-trained weights based on the NIQE metric [28].”
An Experiment-Based Review of Low-Light Image Enhancement Methods (2020, methodology): “Eight metrics, namely, STD, IE, AG, BLIINDS-II [290], NIQE [289], BRISQUE [288], the contrast enhancement-based contrast-changed image quality measure (CEIQ) [311], and the spatial–spectral entropy-based quality measure (SSEQ) [312], were employed for NIQA-based evaluation.”
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.”