No-Reference Image Quality Assessment via Transformers, Relative Ranking, and Self-Consistency

Golestaneh, S. Alireza; Dadsetan, Saba; Kitani, Kris M.

doi:10.1109/wacv51458.2022.00404

Cited by 157 publications

(43 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several no-reference IQA frameworks are proposed based on both the traditional and deep learning methods [ 9 , 26 ]. In this regard, Golestaneh et al [ 27 ] presented a transformer and convolutional neural network (CNN)-based assembled method to rank the images based on local and non-local features. Similarly, Huang et al [ 28 ] proposed a multi-region adjacent pixels correlation (MR-PC) approach to assess the quality of panorama images.…”

Section: Related Workmentioning

confidence: 99%

Subjective Assessment of Objective Image Quality Metrics Range Guaranteeing Visually Lossless Compression

Afnan

Ullah

Yaseen

et al. 2023

Sensors

View full text Add to dashboard Cite

The usage of media such as images and videos has been extensively increased in recent years. It has become impractical to store images and videos acquired by camera sensors in their raw form due to their huge storage size. Generally, image data is compressed with a compression algorithm and then stored or transmitted to another platform. Thus, image compression helps to reduce the storage size and transmission cost of the images and videos. However, image compression might cause visual artifacts, depending on the compression level. In this regard, performance evaluation of the compression algorithms is an essential task needed to reconstruct images with visually or near-visually lossless quality in case of lossy compression. The performance of the compression algorithms is assessed by both subjective and objective image quality assessment (IQA) methodologies. In this paper, subjective and objective IQA methods are integrated to evaluate the range of the image quality metrics (IQMs) values that guarantee the visually or near-visually lossless compression performed by the JPEG 1 standard (ISO/IEC 10918). A novel “Flicker Test Software” is developed for conducting the proposed subjective and objective evaluation study. In the flicker test, the selected test images are subjectively analyzed by subjects at different compression levels. The IQMs are calculated at the previous compression level, when the images were visually lossless for each subject. The results analysis shows that the objective IQMs with more closely packed values having the least standard deviation that guaranteed the visually lossless compression of the images with JPEG 1 are the feature similarity index measure (FSIM), the multiscale structural similarity index measure (MS-SSIM), and the information content weighted SSIM (IW-SSIM), with average values of 0.9997, 0.9970, and 0.9970 respectively.

show abstract

Section: Related Workmentioning

confidence: 99%

Subjective Assessment of Objective Image Quality Metrics Range Guaranteeing Visually Lossless Compression

Afnan

Ullah

Yaseen

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…A meta‐learning approach classifies each distorted version of a tested image. After that, the relevant distorted image representation features are extracted by augmenting the training data through data augmentation (such as image rotation, flipping, grayscale, and visual enhancement) or an auxiliary image processing task that is different but related to the original NR‐IQA task 47 …”

Section: Related Workmentioning

confidence: 99%

“…PaQ2PiQ : The PaQ2PiQ data set contains about 40,000 real‐world, mixed‐distortion, multiscale, multicontent images and about 120,000 cropped image patches of various scales and aspect ratios as an IQA database 47 …”

Section: Performance Evaluationmentioning

confidence: 99%

Contrastive distortion‐level learning‐based no‐reference image‐quality assessment

Wei

Zhou

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

A contrastive distortion-level learning-based no-reference image-quality assessment (NR-IQA) framework is proposed in this study to further effectively model various distortion types with the same or different distortion levels.The proposed method aims to improve the prediction accuracy of NR-IQA. The proposed method consists of three parts: multiscale distortion-level representation learning, single-image NR-IQA, and a representation affinity module, which can reduce NR-IQA computational complexity while maintaining a low-distortion representation of high-distortion inputs. The proposed NR-IQA method aims to extract distributional features of samples in real distorted images and predict ambiguity based on distortion-level learning. Experimental results show that by comparing on many NR-IQA data sets the proposed method can outperform state-of-the-art methods.

show abstract

“…First, as shown in Figure 1, we take the original image xi as the input of branch 1 and take its horizontally flipped version as the input of branch 2. In this way, we can learn stable features and reduce the model's uncertainty caused by the data transformations [12]. Second, as illustrated in Section 3.1.3, we utilize different parameter initialization methods to classifiers in two branches for different prediction abilities in the early training stage.…”

Section: Overview Of Architecturementioning

confidence: 99%

Improving IQA Performance Based on Deep Mutual Learning

Yue

Cheng

et al. 2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

In this paper, we propose a novel solution, termed DML-IQA, for the image quality assessment (IQA) tasks. DML-IQA holds a dual-branch network architecture and builds the IQA model through a deep mutual learning (DML) strategy. Specifically, the two branches extract stable feature representations by feeding different transformed images into the classical CNNs. The DML strategy first calculates the prediction loss of each branch and the consistency loss across two branches, followed by updating the network iteratively to converge. Overall, DML-IQA has the following advantages: 1) It is flexible to adapt to diverse backbones for tackling the IQA issues in both the laboratory and wild; 2) It improves the baseline's performance by approximately 1%~2%, especially performs well in the case of small samples. Extensive experiments on four public datasets show that the proposed DML-IQA can handle the IQA tasks with considerable effectiveness and generalization.

show abstract

No-Reference Image Quality Assessment via Transformers, Relative Ranking, and Self-Consistency

Cited by 157 publications

References 56 publications

Subjective Assessment of Objective Image Quality Metrics Range Guaranteeing Visually Lossless Compression

Subjective Assessment of Objective Image Quality Metrics Range Guaranteeing Visually Lossless Compression

Contrastive distortion‐level learning‐based no‐reference image‐quality assessment

Improving IQA Performance Based on Deep Mutual Learning

Contact Info

Product

Resources

About