Quantitative evaluation of omnidirectional video quality

Birkbeck, Neil; Brown, Chip; Suderman, Robert

doi:10.1109/qomex.2017.7965684

Cited by 7 publications

(14 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Planar metrics. None of the metrics proposed in literature [5][6][7][8][9][10][11] addresses the detection of localized distortions in 360-degree content due to lossy compression in CM domain. Therefore, the performance of the proposed method is compared to that of a set of classical FR quality metrics for 2D images, i.e., planar metrics.…”

Section: Methodsmentioning

confidence: 99%

“…Results. All metrics were computed on a vertically subsampled version of the cube faces, corresponding to a face resolution of 1138x569 pixels 6 . Table 1 reports the detection accuracy achieved by each metric, using the different classification techniques, with L = 4 and L = 8 7 .…”

Section: Methodsmentioning

confidence: 99%

“…Although approaches to quantify the overall visual quality of 360-degree images and videos have been proposed in literature [5][6][7][8][9][10][11], they consider uniformly distributed distortions due to the processing of content in ER domain, rather than localized distortions that might occur when processing the signal in other domains, such as the CM one [2].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Graph-Based Detection of Seams In 360-Degree Images

Simone

Azevedo

Kim

et al. 2019

2019 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

In this paper, we propose an algorithm to detect a specific kind of distortions, referred to as seams, which commonly occur when a 360-degree image is represented in planar domain by projecting the sphere to a polyhedron, e.g, via the Cube Map (CM) projection, and undergoes lossy compression. The proposed algorithm exploits a graph-based representation to account for the actual sampling density of the 360-degree signal in the native spherical domain. The CM image is considered as a signal lying on a graph defined on the spherical surface. The spectra of the processed and the original signals, computed by applying the Graph Fourier Transform, are compared to detect the seams. To test our method a dataset of compressed CM 360-degree images, annotated by experts, has been created. The performance of the proposed algorithm is compared to those achieved by baseline metrics, as well as to the same approach based on spectral comparison but ignoring the spherical nature of the signal. The experimental results show that the proposed method has the best performance and can successfully detect up to approximately 90% of visible seams on our dataset.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Graph-Based Detection of Seams In 360-Degree Images

Simone

Azevedo

Kim

et al. 2019

2019 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

show abstract

“…The proposed approach is based on extracting spatio-temporal quality features (i.e., computing objective IQA metrics) from viewports, temporally pooling them taking the characteristics of the human-visual system (HVS) into consideration, and then training a random forest regression model to predict the 360-degree video quality. On the one hand, working with viewports allow us to better account for the final viewed content and naturally supports different projections [6,7]. On the other hand, the use of multiple objec-tive metrics computed on these viewports allow our method to have a good performance for the complex and diversifying nature of visual distortions appearing in 360-degree videos.…”

Section: Introductionmentioning

confidence: 94%

A Viewport-Driven Multi-Metric Fusion Approach for 360-Degree Video Quality Assessment

Azevedo

Birkbeck²,

Janatra³

et al. 2020

2020 IEEE International Conference on Multimedia and Expo (ICME)

Self Cite

View full text Add to dashboard Cite

We propose a new viewport-based multi-metric fusion (MMF) approach for visual quality assessment of 360-degree (omnidirectional) videos. Our method is based on computing multiple spatio-temporal objective quality metrics (features) on viewports extracted from 360-degree videos, and learning a model that combines these features into a metric that closely matches subjective quality scores. The main motivations for the proposed method are that: 1) quality metrics computed on viewports better captures the user experience than metrics computed on the projection domain; 2) no individual objective image quality metric always performs the best for all types of visual distortions, while a learned combination of them is able to adapt to different conditions. Experimental results, based on the largest available 360-degree videos quality dataset, demonstrate that the proposed metric outperforms state-of-the-art 360-degree and 2D video quality metrics.

show abstract

“…It is shown that most test subjects are focused on the front region of the video while watching the ODVs, which is an expected result. Birkbeck et al in [17] analyse the perceived ODV quality in relation to the type of video projection (Equirectangular, Cubemap, and Equi-Angular Cubemap projection), showing how the analysis of ODV quality and user experience requires devoting more time to additional research activities since the quality of 2D videos can be evaluated without focusing on these aspects. The relationship between ODV projection and the user experience is also analysed in [18,19].…”

Section: Related Workmentioning

confidence: 99%

Subjective Evaluation of User Quality of Experience for Omnidirectional Video Streaming

Mrvelj

Matulin

Martirosov

2020

PROMET

View full text Add to dashboard Cite

This paper reports on the results of subjective testing of user Quality of Experience (QoE) for omnidirectional video (ODV) streaming quality. The test was conducted among 20 test subjects who watched three ODVs using a Head Mounted Display (HMD) system. The length of the videos was between two and three minutes. The first video was used for training purposes and contained no quality degradations. The quality of the other two ODVs was degraded by manipulating the resolution or by introducing different frame drop patterns. While watching the pre-prepared videos the subjects indicated if they noticed the changes in the quality and then rated it. After watching each video, the subjects completed a separate questionnaire, which evaluated their level of enjoyment and discomfort with the video. The results showed that the degradation of both objective parameters (video resolution and frame rate) impacted the subjects’ perception of quality; however, the impact was somewhat alleviated in ODV which contained dynamic scenes and fast camera movements.

show abstract

Quantitative evaluation of omnidirectional video quality

Cited by 7 publications

References 7 publications

Graph-Based Detection of Seams In 360-Degree Images

Graph-Based Detection of Seams In 360-Degree Images

A Viewport-Driven Multi-Metric Fusion Approach for 360-Degree Video Quality Assessment

Subjective Evaluation of User Quality of Experience for Omnidirectional Video Streaming

Contact Info

Product

Resources

About