Achieving high angular resolution via view synthesis: Quality assessment of 3D content on super multiview lightfield display

Super multiview (SMV) lightfield displays provide glasses- free immersive 3D experience via continuous motion parallax. Such displays require input views acquired at high angular resolution, which could be economically and technologically prohibitive. To achieve high view density, synthesizing absent views from the captured ones has recently been suggested. Clearly, a critical proportion of the views should be unsynthesized in order to ensure acceptable quality of the 3D content. In order to determine the said proportion we subjectively assessed the degradation in perceptual 3D quality vis-à- vis the proportion of synthesized views. The qualitative analysis of the subjective scores ascertained that the quality of 3D perception is mainly affected by discontinuities in motion parallax introduced by view synthesis, along with 2D quality of the constituent views. The subjective scores collected in the process exhibited a three-regime behaviour, corresponding to small, moderate and large proportion of synthesized views. For a high 3D quality experience, we recommend operation in the first regime, albeit its high sensitivity to the proportion of synthesized views. The second regime, less susceptible to the said proportion, can be considered when a moderate perceptual quality is acceptable. The third regime, exhibiting monotonous decrease in perceptual quality, is not recommended.

[1]  Rajiv Soundararajan,et al.  Study of Subjective and Objective Quality Assessment of Video , 2010, IEEE Transactions on Image Processing.

[2]  Maria G. Martini,et al.  On the edge of the seat: Reduced angular resolution of a light field cinema with fixed observer positions , 2017, 2017 Ninth International Conference on Quality of Multimedia Experience (QoMEX).

[3]  Neil A. Dodgson,et al.  Three-Dimensional Displays: A Review and Applications Analysis , 2011, IEEE Transactions on Broadcasting.

[4]  Santanu Chaudhury,et al.  A Novel Hybrid Kinect-Variety-Based High-Quality Multiview Rendering Scheme for Glass-Free 3D Displays , 2017, IEEE Transactions on Circuits and Systems for Video Technology.

[5]  C.-C. Jay Kuo,et al.  Subjective and Objective Video Quality Assessment of 3D Synthesized Views With Texture/Depth Compression Distortion , 2015, IEEE Transactions on Image Processing.

[6]  Weisi Lin,et al.  Quality assessment of 3D synthesized images via disoccluded region discovery , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[7]  Frédéric Dufaux,et al.  Subjective evaluation of Super Multi-View compressed contents on high-end light-field 3D displays , 2015 .

[8]  Maria G. Martini,et al.  To interpolate or not to interpolate: Subjective assessment of interpolation performance on a light field display , 2017, 2017 IEEE International Conference on Multimedia & Expo Workshops (ICMEW).

[9]  Sumohana S. Channappayya,et al.  Super-multiview content with high angular resolution: 3D quality assessment on horizontal-parallax lightfield display , 2016, Signal Process. Image Commun..

[10]  Maria G. Martini,et al.  How to rate the new glasses-free experience : subjective quality assessment methodologies for experiments on light field displays , 2017 .

[11]  Takanori Senoh,et al.  New visual coding exploration in MPEG: Super-MultiView and Free Navigation in Free viewpoint TV , 2016, SD&A.

[12]  Frederic Dufaux,et al.  Emerging Technologies for 3D Video: Creation, Coding, Transmission and Rendering , 2013, Emerging Technologies for 3D Video.

[13]  Maria G. Martini,et al.  Perceived quality of angular resolution for light field displays and the validy of subjective assessment , 2016, 2016 International Conference on 3D Imaging (IC3D).

[14]  Attila Barsi,et al.  View synthesis for lightfield displays using region based non-linear image warping , 2012, 2012 International Conference on 3D Imaging (IC3D).

[15]  Patrick Le Callet,et al.  Objective image quality assessment of 3D synthesized views , 2015, Signal Process. Image Commun..