Graphics Reference
In-Depth Information
6.2 Challengesforreal-time3Dvideoqualityevaluation
The possibility to measure 3D image/video quality in real time, as requested by 3D video
applications, is hindered by several issues. The major challenge is how we could measure
theeffectofallperceptualattributes(e.g.,depth,presence,naturalness,etc.)associatedwith
3D viewing. The lack of availability of common 3D image/video databases is also detri-
mental for the advance in this discipline. The following paragraphs briefly discuss these
challenges and possible solutions foreseen.
6.2.1 Measurement of different 3D perceptual attributes
Even though emerging 3D quality evaluation methods accurately predict a given quality
attribute, the relationship among these perception attributes has not be thoroughly studied.
The combined effect directly affects user experience and can be measured using emerging
QoE indices. Therefore the current need is to understand how 3D audio/image processing
and transmission artifacts affect the overall experience of the user, then identify audio,
imageandcontextualfeatureswhichcanbeusedtoquantifythe overalleffectonuserexper-
ience.Ontheotherhand,itisnecessarytounderstandhowtheHVSperceives these 3D ar-
tifacts. For instance, there could be conflicts based on whether binocular suppression or
binocularrivalryistakingplacebasedontheartifactsinquestion.Theseaspectsneedexten-
dedattention in order to measure the overall experience of 3D viewing. In order to enable
a unified approach to 3D objective quality subjective quality evaluation studies, standard-
ization of these procedures are necessary. Several standardization activities are being car-
ried out by VQEG, ITU (Recommendations: ITU-T P- and J-series), European Broad-
castingUnionEBU(3D-TVGroup)andotherStandardsDevelopingOrganizations(SDOs)
in relation to 3D video subjective and objective quality evaluations. Currently, the Video
Quality Expert Group (VQEG) is working (3DTV project) on creating a ground truth 3D
video dataset (GroTruQoE dataset) using the pair-comparison method. This ground truth
database will then be used to evaluate other time-efficient 3D subjective quality evaluation
methodologies and objective quality models. In addition, the project also addresses the
objective quality assessment of 3D video, with the plan to evaluate 3D quality of experi-
ence in relation to the visual quality, depth quality and visual comfort dimensions. Most of
these findings are reported to objective and subjective 3D video quality studies in ITU-T
Study Groups (SG) 9 and 12. EBU is also working on 3D video production, formats and
sequence properties for 3DTV Broadcasting applications (e.g., EBU Recommendation R
135).
