From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality

Ying, Zhenqiang; Niu, Haoran; Gupta, P. K. Sen; Mahajan, Dhruv; Ghadiyaram, Deepti; Bovik, Alan C.

doi:10.1109/cvpr42600.2020.00363

Cited by 166 publications

(109 citation statements)

References 51 publications

Supporting

Mentioning

107

Contrasting

Order By: Relevance

“…Several assessment indices are used to compare SOTA dehazing models, namely the peak signal-to-noise ratio (PSNR) and the structural similarity index (SSIM) for evaluations based on reference haze-free images, the average gradient (AG) [ 15 ], image entropy (IE) [ 35 ], the fog-aware density evaluator (FADE) [ 33 ] provided with the LIVE dataset, blind image quality measure of enhanced images (BIQME) [ 36 ], and patches-to-pictures quality predictor (PaQ2PiQ) [ 37 ] for non-reference image quality evaluation. AG is related to the edges and variance in an image and is defined by (7).…”

Section: Resultsmentioning

confidence: 99%

“…BIQME [ 36 ] extracts 17 features by analyzing contrast, sharpness, brightness etc., and it predicts evaluations on image quality by training a regression module. Meanwhile, PaQ2PiQ [ 37 ] evaluates both global and local image quality by training a modified ResNet-18 [ 28 ] on a new dataset. This training dataset comprises pictures and patches associated with human perceptual quality judgements, i.e., mean opinion score (MOS).…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Lightweight and Efficient Image Dehazing Network Guided by Transmission Estimation from Real-World Hazy Scenes

Zhan

Zhang

Zhong

et al. 2021

Sensors

View full text Add to dashboard Cite

In this paper, a transmission-guided lightweight neural network called TGL-Net is proposed for efficient image dehazing. Unlike most current dehazing methods that produce simulated transmission maps from depth data and haze-free images, in the proposed work, guided transmission maps are computed automatically using a filter-refined dark-channel-prior (F-DCP) method from real-world hazy images as a regularizer, which facilitates network training not only on synthetic data, but also on natural images. A double-error loss function that combines the errors of a transmission map with the errors of a dehazed image is used to guide network training. The method provides a feasible solution for introducing priors obtained from traditional non-learning-based image processing techniques as a guide for training deep neural networks. Extensive experimental results demonstrate that, in terms of several reference and non-reference evaluation criteria for real-world images, the proposed method can achieve state-of-the-art performance with a much smaller network size and with significant improvements in efficiency resulting from the training guidance.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Lightweight and Efficient Image Dehazing Network Guided by Transmission Estimation from Real-World Hazy Scenes

Zhan

Zhang

Zhong

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Deep convolutional neural networks (CNNs) have been shown to deliver standout performance in a wide variety of low-level computer vision applications [17], [23], [25], [58]. Recently, the release of several large-scale psychometric visual quality databases [29]- [32], [51] have sped the application of deep CNNs to perceptual video and image quality modeling.…”

Section: B Deep Learning-based Bvqa Modelsmentioning

confidence: 99%

“…CNN-based solutions have been observed to generally perform well on UGC picture quality problems [17], [27], [51] thanks to several recently released large-scale picture quality datasets [17], [51], [82]. Still, none of them have proven effective on UGC video quality databases [30]- [32].…”

Section: Deep Learning Featuresmentioning

confidence: 99%

RAPIQUE: Rapid and Accurate Video Quality Prediction of User Generated Content

Wang

et al. 2021

IEEE Open J. Signal Process.

104

View full text Add to dashboard Cite

Blind or no-reference video quality assessment of user-generated content (UGC) has become a trending, challenging, heretofore unsolved problem. Accurate and efficient video quality predictors suitable for this content are thus in great demand to achieve more intelligent analysis and processing of UGC videos. Previous studies have shown that natural scene statistics and deep learning features are both sufficient to capture spatial distortions, which contribute to a significant aspect of UGC video quality issues. However, these models are either incapable or inefficient for predicting the quality of complex and diverse UGC videos in practical applications. Here we introduce an effective and efficient video quality model for UGC content, which we dub the Rapid and Accurate Video Quality Evaluator (RAPIQUE), which we show performs comparably to state-of-the-art (SOTA) models but with orders-of-magnitude faster runtime. RAPIQUE combines and leverages the advantages of both quality-aware scene statistics features and semantics-aware deep convolutional features, allowing us to design the first general and efficient spatial and temporal (space-time) bandpass statistics model for video quality modeling. Our experimental results on recent large-scale UGC video quality databases show that RAPIQUE delivers top performances on all the datasets at a considerably lower computational expense. We hope this work promotes and inspires further efforts towards practical modeling of video quality problems for potential real-time and low-latency applications. To promote public usage, an implementation of RAPIQUE has been made freely available online: https://github.com/vztu/RAPIQUE.

show abstract

“…Apart from MOSs, EXIF data, image attributes, and scene category labels were also recorded to facilitate the development of BIQA models for real-world applications. Concurrently, Ying et al [23] built a large dataset that contains patch quality annotations. As discussed previously, different datasets may use different subjective procedures, leading to different perceptual scales of the collected MOSs.…”

Section: Iqa Datasetsmentioning

confidence: 99%

Learning to Rank for Blind Image Quality Assessment

Zheng

Zhang

Cai

2020

Proceedings of the 4th International Conference on Computer Science and Application Engineering

View full text Add to dashboard Cite

The explosive growth of image data facilitates the fast development of image processing and computer vision methods for emerging visual applications, meanwhile introducing novel distortions to the processed images. This poses a grand challenge to existing blind image quality assessment (BIQA) models, failing to continually adapt to such subpopulation shift. Recent work suggests training BIQA methods on the combination of all available human-rated IQA datasets. However, this type of approach is not scalable to a large number of datasets, and is cumbersome to incorporate a newly created dataset as well. In this paper, we formulate continual learning for BIQA, where a model learns continually from a stream of IQA datasets, building on what was learned from previously seen data. We first identify five desiderata in the new setting with a measure to quantify the plasticity-stability trade-off. We then propose a simple yet effective method for learning BIQA models continually. Specifically, based on a shared backbone network, we add a prediction head for a new dataset, and enforce a regularizer to allow all prediction heads to evolve with new data while being resistant to catastrophic forgetting of old data. We compute the quality score by an adaptive weighted summation of estimates from all prediction heads. Extensive experiments demonstrate the promise of the proposed continual learning method in comparison to standard training techniques for BIQA.

show abstract

From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality

Cited by 166 publications

References 51 publications

Lightweight and Efficient Image Dehazing Network Guided by Transmission Estimation from Real-World Hazy Scenes

Lightweight and Efficient Image Dehazing Network Guided by Transmission Estimation from Real-World Hazy Scenes

RAPIQUE: Rapid and Accurate Video Quality Prediction of User Generated Content

Learning to Rank for Blind Image Quality Assessment

Contact Info

Product

Resources

About