Which Has Better Visual Quality: The Clear Blue Sky or a Blurry Animal?

Li, Dingquan; Jiang, Tingting; Wang, Lin; Jiang, Ming

doi:10.1109/tmm.2018.2875354

Cited by 107 publications

(54 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Content-aware features can help addressing content-dependency on the predicted image/video quality, so as to improve the performance of objective models [13,17,41,49]. Jaramillo et al [13] Figure 2: The overall framework of the proposed method.…”

Section: Content-aware Featuresmentioning

confidence: 99%

“…Siahaan et al [41] and Wu et al [49] utilize semantic information from the top layer of pre-trained image classification networks to incorporate with traditional quality features. Li et al [17] exploit the deep semantic feature aggregation of multiple patches for image quality assessment. It is shown that these deep semantic features alleviate the impact of content on the quality assessment task.…”

Section: Modeling Of Temporal-memory Effects Content-aware Feature Exmentioning

confidence: 99%

“…Inspired by their work, we consider using pretrained image classification networks for content-aware feature extraction as well. Unlike the work in [17], to get the features, we directly feed the whole frame into the network and apply not only global average pooling but also global standard deviation pooling to the output semantic feature maps. Since our work aims at the VQA task, we further put forward a new module for modeling temporal characteristics of human behaviors when rating video quality.…”

Section: Modeling Of Temporal-memory Effects Content-aware Feature Exmentioning

confidence: 99%

See 2 more Smart Citations

Quality Assessment of In-the-Wild Videos

Jiang

2019

Proceedings of the 27th ACM International Conference on Multimedia

Self Cite

240

245

View full text Add to dashboard Cite

Quality assessment of in-the-wild videos is a challenging problem because of the absence of reference videos and shooting distortions. Knowledge of the human visual system can help establish methods for objective quality assessment of in-the-wild videos. In this work, we show two eminent effects of the human visual system, namely, content-dependency and temporal-memory effects, could be used for this purpose. We propose an objective no-reference video quality assessment method by integrating both effects into a deep neural network. For content-dependency, we extract features from a pre-trained image classification neural network for its inherent content-aware property. For temporal-memory effects, long-term dependencies, especially the temporal hysteresis, are integrated into the network with a gated recurrent unit and a subjectivelyinspired temporal pooling layer. To validate the performance of our method, experiments are conducted on three publicly available inthe-wild video quality assessment databases: KoNViD-1k, CVD2014, and LIVE-Qualcomm, respectively. Experimental results demonstrate that our proposed method outperforms five state-of-the-art methods by a large margin, specifically, 12.39%, 15.71%, 15.45%, and 18.09% overall performance improvements over the second-best method VBLIINDS, in terms of SROCC, KROCC, PLCC and RMSE, respectively. Moreover, the ablation study verifies the crucial role of both the content-aware features and the modeling of temporalmemory effects. The PyTorch implementation of our method is released at https://github.com/lidq92/VSFA. KEYWORDSvideo quality assessment; human visual system; content dependency; temporal-memory effects; in-the-wild videos ACM Reference Format:

show abstract

Section: Content-aware Featuresmentioning

confidence: 99%

Section: Modeling Of Temporal-memory Effects Content-aware Feature Exmentioning

confidence: 99%

Section: Modeling Of Temporal-memory Effects Content-aware Feature Exmentioning

confidence: 99%

See 1 more Smart Citation

Quality Assessment of In-the-Wild Videos

Jiang

2019

Proceedings of the 27th ACM International Conference on Multimedia

Self Cite

240

245

View full text Add to dashboard Cite

show abstract

“…A more realistic way to evaluate the output quality of enhancement algorithms is to use computational models, which can automatically predict the quality of images as perceived by humans. These objective metrics can be classified into full-reference (FR) ( [37], [38]), reducedreference (RR) ( [39], [40], [41]), and no-reference (NR) ( [42], [43], [44], [45], [46], [47], [48]) approaches, depending on whether the original image is used as the reference of highest quality [37], [49], [50]. But in image enhancement the original image does not represent such reference of perfect/maximum quality as an enhanced image may be of higher or lower quality relative to the original.…”

Section: Objective Evaluation Of Visual Qualitymentioning

confidence: 99%

Underwater Image Quality Assessment: Subjective and Objective Methods

Guo

Liu

et al. 2022

IEEE Trans. Multimedia

View full text Add to dashboard Cite

Underwater image enhancement plays a critical role in marine industry. Various algorithms are applied to enhance underwater images, but their performance in terms of perceptual quality has been little studied. In this paper, we investigate five popular enhancement algorithms and their output image quality. To this end, we have created a benchmark, including images enhanced by different algorithms and ground truth image quality obtained by human perception experiments. We statistically analyse the impact of various enhancement algorithms on the perceived quality of underwater images. Also, the visual quality provided by these algorithms is evaluated objectively, aiming to inform the development of objective metrics for automatic assessment of the quality for underwater image enhancement. The image quality benchmark and its objective metric are made publicly available.

show abstract

“…Unfortunately, MPI-LFA dataset does not provide enough statistical information about the results rather than the JOD scores, we utilized the cross-validation methodology for the evaluation of the proposed model Cross-validation is another alternative to reliably evaluate the generalization performance of a machine-learning-based quality assessment system [19]. Therefore, the proposed model was evaluated through 1000-fold cross-validation as done in [14,20,21]. At each fold, the whole dataset was randomly divided into 80% for training and 20% for testing, without overlap between them for 1000 times.…”

Section: Experimental Setup and Evaluation Criteriamentioning

confidence: 99%

No-Reference Quality Evaluation of Light Field Content Based on Structural Representation of The Epipolar Plane Image

Ak¹,

Ling²,

Callet³

2020

2020 IEEE International Conference on Multimedia &Amp; Expo Workshops (ICMEW)

View full text Add to dashboard Cite

As an emerging technology, Light Field (LF) has gained everincreasing importance in the domain. In order to provide guidance for the development of perceptually accurate Light Field (LF) processing tools and supervise the entire streaming system, robust perceptual quality assessment metrics are required. Especially, No-Reference (NR) metrics are preferable to compare LF with different angular resolutions. Some metrics have been developed by extending commonly used 2D image quality metrics to the 4D LF domain with angular consistency terms. Nonetheless, although these models consistently show slightly improved performance, most of them are limited in evaluating the quality of LF using the sub-aperture views with additional terms on the angular domain. There is an evident lack of reliable quality metrics that are tailored to LF content. To remedy this lack, we propose a NR quality metric for LF contents based on representing EPI with structural descriptors, including the Histogram of Gradients and the Convolutional Sparse Coding based descriptors. The primary motivation resides in our observations that (1) LF related distortions on the angular domain are highly noticeable on the Epipolar Plane Image representations (EPI); (2) most of the distortions in EPI are structurerelated. Extensive experiments on the MPI-LFA [1] LF image quality dataset demonstrate that our method provides competitive performance with the state-of-the-art NR image quality metrics.

show abstract

Which Has Better Visual Quality: The Clear Blue Sky or a Blurry Animal?

Cited by 107 publications

References 57 publications

Quality Assessment of In-the-Wild Videos

Quality Assessment of In-the-Wild Videos

Underwater Image Quality Assessment: Subjective and Objective Methods

No-Reference Quality Evaluation of Light Field Content Based on Structural Representation of The Epipolar Plane Image

Contact Info

Product

Resources

About