Research Article

Open Profiling of Quality: A Mixed Method Approach to Understanding Multimodal Quality Perception

Table 8

Summary of results of three experiments using the open profiling of quality.

Experiment
Goal: influence of audiovisual depth on perceived quality

Psychoperceptual evaluation: nonsignificant influence
Sensorial profiling: impressions of content, visual presentation mode and room acoustic
Visual presentation mode: 2D described as error-free while 3D mainly as erroneousness, some positive mentions (descripstions of 3D, spacious, tangible).
 Audio: impressions of room space, divided according to room acoustic models
Participants’ sensorial preferences towards audio, video, or both modalities
External preference mapping: N/A

Experiment
Goal: influence of audiovisual depth on experienced quality of mobile 3D television and video

Psychoperceptual evaluation: dominance of visual quality factors over audio and audiovisual quality
Sensorial profiling: impressions of visual presentation mode and affective factors
Visual presentation mode: 2D—neutral/positive quality descriptions (pleasant, beautiful, focusable);
 3D—positive (spatial, illusory, layered, depth impression), negative (artifacts, flickering, stressful, blurred)
 Experienced added value of depth is visible only if the level of artifacts is low
External preference mapping: one content is clearly preferred in 3D described as spatial, to come forth, and three-dimensional while others in 2D described as interesting, stress-free, pleasant

Experiment
Goal: influence of video coding methods on perceived quality of mobile 3D television and video

Psychoperceptual evaluation: the most satisfying quality provided by multiview coding and video + depth coding methods when averaged over the contents; some content-dependencies exist
Sensorial profiling: added value of depth conveyed when level of artifacts is low
 Visual 3D—positive quality (sharp, detailed, resolution) versus negative quality (blurry, blocky grainy)
External preference mapping: artifact-free videos preferred; exceptional contents identified