VQone MATLAB toolbox: A graphical experiment builder for image and video quality evaluations

Nuutinen, Mikko; Virtanen, Toni; Rummukainen, Olli; Häkkinen, Jukka

doi:10.3758/s13428-014-0555-y

Cited by 12 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have recruited 10 viewers, a number that is considered enough for subjective image quality evaluation tests [35]; none of the viewers was familiar with image quality evaluation or the work presented in this paper. Evaluation has been done following a DSIS (Double-Stimulus Impairment Scale) setup, created using VQone, a tool specifically designed for this type of experiments [25]: subjects evaluated the test image in comparison to the original image, and graded how similar is the test image to the original, using a continuous scale from 0 to 100, with no marked values to avoid choosing preferred numbers. We have randomly selected 50 images from the BSD500 dataset, containing different subjects, such as nature scenes, man-made objects, persons, animals, etc.…”

Section: Subjective Evaluationmentioning

confidence: 99%

Deep Generative Adversarial Compression Artifact Removal

Galteri

Seidenari

Bertini

et al. 2017

2017 IEEE International Conference on Computer Vision (ICCV)

192

View full text Add to dashboard Cite

Compression artifacts arise in images whenever a lossy compression algorithm is applied. These artifacts eliminate details present in the original image, or add noise and small structures; because of these effects they make images less pleasant for the human eye, and may also lead to decreased performance of computer vision algorithms such as object detectors. To eliminate such artifacts, when decompressing an image, it is required to recover the original image from a disturbed version. To this end, we present a feed-forward fully convolutional residual network model trained using a generative adversarial framework. To provide a baseline, we show that our model can be also trained optimizing the Structural Similarity (SSIM), which is a better loss with respect to the simpler Mean Squared Error (MSE).Our GAN is able to produce images with more photorealistic details than MSE or SSIM based networks. Moreover we show that our approach can be used as a pre-processing step for object detection in case images are degraded by compression to a point that state-of-the art detectors fail. In this task, our GAN method obtains better performance than MSE or SSIM trained networks.

show abstract

Section: Subjective Evaluationmentioning

confidence: 99%

Deep Generative Adversarial Compression Artifact Removal

Galteri

Seidenari

Bertini

et al. 2017

2017 IEEE International Conference on Computer Vision (ICCV)

192

View full text Add to dashboard Cite

show abstract

“…There were no missing data in the preference evaluations, nor were there inaccurate values outside the requested 0-to-10 mean opinion score (MOS) scale as observers’ preferences were recorded using a VQOne MATLAB program that prevented omission of answers or using inappropriate values outside the given scale (Nuutinen et al, 2016). Figure 4 shows that the preference MOS follows normal distribution quite well.…”

Section: Resultsmentioning

confidence: 99%

“…The durations include vision tests, instructions, practice rounds, and breaks. All experiments were conducted using the VQone MATLAB toolbox, a graphical experiment builder for image and video quality evaluations (Nuutinen, Virtanen, Rummukainen, & Häkkinen, 2016).…”

Section: Methodsmentioning

confidence: 99%

Underlying elements of image quality assessment: Preference and terminology for communicating image quality characteristics.

Virtanen¹,

Nuutinen²,

Häkkinen³

2022

Psychology of Aesthetics, Creativity, and the Arts

Self Cite

View full text Add to dashboard Cite

Image quality markedly affects the evaluation of images, and its control is crucial in studies using natural visual scenes as stimuli. Various image elements, such as sharpness or naturalness, can impact how observers view images and more directly how they evaluate their quality. To gain a better understanding of the types of interactions between these various elements, we conducted a study with a large set of images with multiple overlapping distortions, covering a wide range of quality variation. Observers assigned a quality rating on a 0-10 scale plus a verbal description of the images, explaining the elements on which their rating was based. Regression model predicting image quality ratings using 68 attributes uncovered the link between verbal descriptions and quality ratings and the importance of the image quality rating for each of the 68 image attributes. Brightness, naturalness, and good colors seem to be related to the highest image quality preference. However, the most important elements for predicting good image quality were related to image fidelity such as graininess and sharpness. This indicates that a certain level of image fidelity must be achieved before more subjective associations with, for instance, naturalness can emerge. Of the attributes, 72% had a negative impact on the preference judgment. This negative bias may be due to the fact that there are more ways that observers can perceive an image to fail than to excel when they are asked to evaluate image quality.

show abstract

“…All experiments in the display photograph studies were conducted using the VQone MATLAB toolbox. 42 Altogether 62 different scenes were collected for the studies. The scenes were intended to represent typical photographs that consumers might capture with their camera devices.…”

Section: Materials and Experimental Setupmentioning

confidence: 99%

Image quality wheel

Virtanen

Nuutinen

2019

J. Electron. Imag.

Self Cite

View full text Add to dashboard Cite

We have collected a large dataset of subjective image quality "*nesses," such as sharpness or colorfulness. The dataset comes from seven studies and contains 39,415 quotations from 146 observers who have evaluated 62 scenes either in print images or on display. We analyzed the subjective evaluations and formed a hierarchical image quality attribute lexicon for *nesses, which is visualized as image quality wheel (IQ-Wheel). Similar wheel diagrams for attributes have become industry standards in other sensory experience fields such as flavor and fragrance sciences. The IQ-Wheel contains the frequency information of 68 attributes relating to image quality. Only 20% of the attributes were positive, which agrees with previous findings showing a preference for negative attributes in image quality evaluation. Our results also show that excluding physical attributes of paper gloss, observers then use similar terminology when evaluating images with printed images or images viewed on a display. IQ-Wheel can be used to guide the selection of scenes and distortions when designing subjective experimental setups and creating image databases.

show abstract

VQone MATLAB toolbox: A graphical experiment builder for image and video quality evaluations

Cited by 12 publications

References 20 publications

Deep Generative Adversarial Compression Artifact Removal

Deep Generative Adversarial Compression Artifact Removal

Underlying elements of image quality assessment: Preference and terminology for communicating image quality characteristics.

Image quality wheel

Contact Info

Product

Resources

About