Graphics Reference
In-Depth Information
6 Qualityevaluationof3Dvideo
Yourgoalsforthis“Qualityevaluationof3Dvideo”chapteraretolearnabout:
• Perceptual quality evaluation of 3D video.
• Subjective 3D video quality evaluation methods.
• Objective 3D video quality evaluation methods.
• Real-time 3D video quality evaluation methods.
Even though the initial developments for 3D video services are in place, the acceptance of
these services is dependent on the user satisfaction of the reconstructed 3D video quality.
Therefore, extensive quality evaluation studies are necessary to study the effect of cam-
era arrangement, data representation, coding, transmission and display techniques on the
perceived quality of 3D video. Some of the stereoscopic image impairments introduced by
the 3D video system are keystone distortion, depth-plane curvature, crosstalk, size distor-
tions, cardboard effect, picket fence effect, image flipping and shear distortion. Moreover,
depending on the coding approaches (e.g. DCT) being used, conventional coding impair-
ments likeblockiness,blurwillbeintroducedtothereconstructed3Dvideo.Theseimpair-
ments in stereoscopic video will influence multi-dimensional perceptual attributes such
as image quality, depth perception, presence, naturalness, etc. A detailed analysis is ne-
cessary to study how these 3D percepts influence the overall perceived quality in gen-
eral. For instance, the study presented in [100] concludes that excessive disparities can
cause eye strain and therefore degrade the perceived image quality. Mostly psychophysical
experiments are conducted to measure and quantify 3D perceptual attributes. In addition to
that, explorative studies can be utilized to get unbiased attitudes and views for emerging
technologieslike3Dvideo.Forinstance,focusgroupscanbeformedtoevaluatetheimpact
of new stereoscopic image systems through group discussions
[101].
This method also can
be employed to evaluate the added value of depth. Moreover, explorative studies will help
better understanding the attributes of a multi- dimensional construct like image quality,
depth perception, viewing experience, etc.
Psychophysical scaling paradigms can be classified into two main categories
[102],
namely;
• Performance-orientedmethods
• Appreciationorientedmethods.
