To avoid the ghosting effect of the light field rendering caused by the insufficient sampling rate, we extend the application of the criterion of ghost-free reconstruction from one constant-depth scenes to more complex scenes which contain multiple depth layers. It is shown that the optimal constant depth and the maximum camera interval can be determined by the scene geometry and the camera resolution. The relationship between them is formulated in this paper. Also, we use an experiment to verify the criterion presented here. The mean square difference (MSD) between the reconstructed views and the standard views are calculated to show the reconstruction quality at different camera intervals. The quantitative data are basically in accord with the subjective observation in this experiment and the results sufficiently support the theoretical analysis.

Date Published: 6 August 2009
PDF: 8 pages
Proc. SPIE 7384, International Symposium on Photoelectronic Detection and Imaging 2009: Advances in Imaging Detectors and Applications, 73841S (6 August 2009); doi: 10.1117/12.835683

Published in SPIE Proceedings Vol. 7384:
International Symposium on Photoelectronic Detection and Imaging 2009: Advances in Imaging Detectors and Applications
Kun Zhang; Xiang-jun Wang; Guang-jun Zhang; Ke-cong Ai, Editor(s)