Multidimensional assessment and principal component analysis of QoE in interactive Multi-View Video and Audio IP communications

This paper compares interactive Multi-View Video and Audio (MVV-A) IP communications and interactive Single-View Video and Audio (SVV-A) ones in terms of QoE. We employ a task which makes use of characteristics of interactive MVV-A and perform a subjective experiment. We conduct multidimensional assessment and principal component analysis of QoE with 14 pairs of polar terms by the SD (Semantic Differential) method, which can assess an object from many viewpoints. We also assess the efficiency of the task objectively. As a result, we see that the users are more satisfied with MVV-A than SVV-A owing to viewpoint change. In addition, we show that the overall satisfaction improves when we select proper play out buffering time. From the result of the principal component analysis, we find that the first principal component is audiovisual quality and user´s feeling, and the second principal component is viewpoint change.