Multi-Pooled Inception Features for No-Reference Image Quality Assessment

Varga, Domonkos

doi:10.3390/app10062186

Cited by 37 publications

(65 citation statements)

References 55 publications

(170 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly to the method of Bianco et al [ 22 ], the layer-wise feature vectors were mapped onto subscores with a trained support vector regressor and the average of the subscores was taken to get the perceptual quality. In [ 24 ], deep features were extracted from multiple Inception modules of pretrained CNNs, concatenated together, and mapped onto quality scores.…”

Section: Introductionmentioning

confidence: 99%

“…Unlike other deep architectures [ 18 , 22 , 23 ], a multi-scale orderless pooling of deep features is elaborated where feature extraction is performed beginning from local random image patches at multiple scales. Unlike our previous method [ 24 ], the focus is on constructing an architecture that extracts deep features from multiple scales of an image rather than examining the effects of deep features extracted from multiple layers of a deep CNN. Extensive experiments have been carried on three large benchmark IQA databases (LIVE In the Wild [ 34 ], KonIQ-10k [ 1 ], and SPAQ [ 2 ]) to demonstrate that the proposed method is able to outperform the state-of-the-art.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

No-Reference Image Quality Assessment with Multi-Scale Orderless Pooling of Deep Features

Varga¹

2021

J. Imaging

Self Cite

View full text Add to dashboard Cite

The goal of no-reference image quality assessment (NR-IQA) is to evaluate their perceptual quality of digital images without using the distortion-free, pristine counterparts. NR-IQA is an important part of multimedia signal processing since digital images can undergo a wide variety of distortions during storage, compression, and transmission. In this paper, we propose a novel architecture that extracts deep features from the input image at multiple scales to improve the effectiveness of feature extraction for NR-IQA using convolutional neural networks. Specifically, the proposed method extracts deep activations for local patches at multiple scales and maps them onto perceptual quality scores with the help of trained Gaussian process regressors. Extensive experiments demonstrate that the introduced algorithm performs favorably against the state-of-the-art methods on three large benchmark datasets with authentic distortions (LIVE In the Wild, KonIQ-10k, and SPAQ).

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

No-Reference Image Quality Assessment with Multi-Scale Orderless Pooling of Deep Features

Varga¹

2021

J. Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

“…Namely, they extracted features with the help of a ResNet [ 27 ] architecture and elaborated a probabilistic representation of distorted images. In [ 28 ], an Inception-V3 [ 29 ] network was utilized as feature extracted and it was pointed out that considering the features of multiple layers is able to improve the performance of perceptual quality prediction. In contrast, Liu et al [ 30 ] trained a Siamese CNN to rank images in terms of perceptual quality.…”

Section: Introductionmentioning

confidence: 99%

No-Reference Image Quality Assessment with Global Statistical Features

Varga

2021

J. Imaging

Self Cite

View full text Add to dashboard Cite

The perceptual quality of digital images is often deteriorated during storage, compression, and transmission. The most reliable way of assessing image quality is to ask people to provide their opinions on a number of test images. However, this is an expensive and time-consuming process which cannot be applied in real-time systems. In this study, a novel no-reference image quality assessment method is proposed. The introduced method uses a set of novel quality-aware features which globally characterizes the statistics of a given test image, such as extended local fractal dimension distribution feature, extended first digit distribution features using different domains, Bilaplacian features, image moments, and a wide variety of perceptual features. Experimental results are demonstrated on five publicly available benchmark image quality assessment databases: CSIQ, MDID, KADID-10k, LIVE In the Wild, and KonIQ-10k.

show abstract

“…Varga [70] introduced no-reference IQM using multi-level inception features from a pretrained CNN. The method uses the entire to extract image resolution independent features.…”

Section: Introductionmentioning

confidence: 99%

Image Quality Assessment without Reference by Combining Deep Learning-Based Features and Viewing Distance

Chetouani

Pedersen

2021

Applied Sciences

View full text Add to dashboard Cite

An abundance of objective image quality metrics have been introduced in the literature. One important essential aspect that perceived image quality is dependent on is the viewing distance from the observer to the image. We introduce in this study a novel image quality metric able to estimate the quality of a given image without reference for different viewing distances between the image and the observer. We first select relevant patches from the image using saliency information. For each patch, a feature vector is extracted from a convolutional neural network model and concatenated at the viewing distance, for which the quality is predicted. The resulting vector is fed to fully connected layers to predict subjective scores for the considered viewing distance. The proposed method was evaluated using the Colourlab Image Database: Image Quality and Viewing Distance-changed Image Database. Both databases provide subjective scores at two different viewing distances. In the Colourlab Image Database: Image Quality we obtain a Pearson correlation of 0.87 at both 50 cm and 100 cm viewing distances, while in the Viewing Distance-changed Image Database we obtained a Pearson correlation of 0.93 and 0.94 at viewing distance of four and six times the image height. The results show the efficiency of our method and its generalization ability.

show abstract

Multi-Pooled Inception Features for No-Reference Image Quality Assessment

Cited by 37 publications

References 55 publications

No-Reference Image Quality Assessment with Multi-Scale Orderless Pooling of Deep Features

No-Reference Image Quality Assessment with Multi-Scale Orderless Pooling of Deep Features

No-Reference Image Quality Assessment with Global Statistical Features

Image Quality Assessment without Reference by Combining Deep Learning-Based Features and Viewing Distance

Contact Info

Product

Resources

About