Segmentation-driven perceptual quality metrics

Cavallaro, Andrea

doi:10.1109/icip.2004.1421881

Cited by 20 publications

(15 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The cognitive behavior of people when watching video sequences has to be taken in consideration as well, as argued by Cavallaro et al. [4] and Bajcsy [5]. A generalization is almost impossible given the large variety of behaviors generated in individuals by similar situations; however, some important characteristics can be extracted.…”

Section: Omnidirectional Video Quality Metricsmentioning

confidence: 98%

Quality Assessment of Panorama Video for Videoconferencing Applications

Leorin

Lucchese

Cutler

2005

2005 IEEE 7th Workshop on Multimedia Signal Processing

View full text Add to dashboard Cite

New video-conference devices based on omnidirectional multi-camera systems have been emerging in the last few years. These devices require innovative and automated video quality assessment in the earlier stages oftheir design in order to guarantee competitive product development and quality monitoring. Current quality assessment techniques are not adequate since they are mostly tailored to single video cameras. Even ifthese techniques are capable of assessing the quality of each video stream separately, the overall quality of a composite video stream generated with the outputs of multiple cameras stitched together presents strong deviations from the results of subjective quality tests. In this paper, we present new strategies for assessing the quality of composite video streams with specific emphasis on the following problems: noticeable calibration differences between adjacent cameras, concentration of motion in limited regions of the panoramic scene, combined vignetting problems, non-uniformity of the surfaces in the seam region between two adjacent cameras. Combining different features from high and low level vision, we evaluate the proposed perceptual quality metric using a set of customized test sequences and verify its correlation with subjective quality tests.

show abstract

Section: Omnidirectional Video Quality Metricsmentioning

confidence: 98%

Quality Assessment of Panorama Video for Videoconferencing Applications

Leorin

Lucchese

Cutler

2005

2005 IEEE 7th Workshop on Multimedia Signal Processing

View full text Add to dashboard Cite

show abstract

“…However, in almost all cases, the presence of a human face in the scene has to be considered extremely important. Accounting for it generally leads to a higher correlation between the quality prediction and the subjective ratings [5]. To detect human faces, we make two sorts of considerations.…”

Section: Face Detectionmentioning

confidence: 99%

“…Following a criterion used, e.g., in [5], two classes of perceptual quality metrics can be distinguished: the first makes use of a general model of the low-level visual processes taking place in the retina and the early visual cortex [4]. Metrics in this class typically require access to the original video or to some features extracted from it, for difference analysis.…”

Section: Introductionmentioning

confidence: 99%

No-reference measurement of perceptually significant blurriness in video frames

Dardi

Abate

Ramponi

2011

SIViP

View full text Add to dashboard Cite

A new method to assess the presence and the strength of the blurring artifact in video frames, without using a reference ideal image, is presented. The estimation is performed first through a global and simple measure over the whole picture, then through a finer, local analysis of the sharpness of the objects borders. The subjective relevance of the scene content is accounted for in choosing the image parts where this local blurriness measurement is performed. Relevant parts are selected using an existing measurement of the saliency of each pixel as well as a simple method to detect human faces, which are a particularly attracting content for the viewer. In the so selected parts, an objective measurement of the blurriness artifact is performed, based on the presence of fine detail and on the local edge width. Then the scene activity, or clutter, is measured by counting the number of distinct picture regions. In an active scene, indeed, a blurred object is deemed to be less apparent. The combination of quantified objective blurriness with clutter yields a final index, which proved to appropriately reproduce the judgements of human observers on images of different quality, resulting from different video encodings.

show abstract

“…Osberger et al proposed a quality metric for the assessment of still pictures, which includes an attention model to weight the influence of visible errors produced by an early vision model of the HVS [8]. In [9], a quality metric which is based on object tracking and segmentation has been introduced and its validity was tested in the case of face segmentation. In [10], a quality metric is proposed for monitoring the quality of video sequences transferred over mobile networks for sign language conversations; since in this application face and hands are the areas on which the user's attention is mostly focused, the distortions in these regions are optimally weighted to create an objective intelligibility score for a distorted sequence.…”

Section: Introductionmentioning

confidence: 99%

Influence of audio-visual attention on perceived quality of standard definition multimedia content

Lee

Simone

Ebrahimi

2009

2009 International Workshop on Quality of Multimedia Experience

View full text Add to dashboard Cite

When human subjects assess the quality of multimedia data, high level perceptual processes such as Focus of Attention (FoA) and eye movements are believed to play an important role in such tasks. While prior art reports incorporation of visual FoA into objective quality metrics, audio-visual FoA has been rarely addressed and utilized in spite of the importance and presence of both audio and video information in many multimedia systems. This paper explores the influence of audio-visual FoA in the perceived quality of standard definition audio-visual sequences. Results of a subjective quality assessment study are reported, where it is shown that the sound source attracts visual attention and thereby the visual degradation in the regions far from the source is less perceived when compared to sound-emitting regions.Index Terms-quality assessment, audio-visual focus of attention, cross-modal interaction, perceived quality

show abstract

Segmentation-driven perceptual quality metrics

Cited by 20 publications

References 17 publications

Quality Assessment of Panorama Video for Videoconferencing Applications

Quality Assessment of Panorama Video for Videoconferencing Applications

No-reference measurement of perceptually significant blurriness in video frames

Influence of audio-visual attention on perceived quality of standard definition multimedia content

Contact Info

Product

Resources

About