Translator Disclaimer
5 June 2014 Bayesian estimation of depth information in three-dimensional integral imaging
Author Affiliations +
In this paper, we propose a Bayesian framework to infer depths of object surfaces in a 3D integral imaging system. In a 3D integral imaging system, the depth of Lambertian surfaces can be estimated from the statistics of the spectral radiation pattern. However, the estimated depth may contain errors due to system uncertainties. To better infer the depth information, we utilize a Bayesian framework and a Markov Random Field (MRF) model with the knowledge of the statistical information of object intensities and the assumption that object surfaces are smooth. In the proposed method, we combine a Bayesian framework and the characteristics of 3D integral imaging systems to infer the depths. Simulated and experimental results illustrate the performance of the proposed method.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiao Xiao, Bahram Javidi, and Dipak K. Dey "Bayesian estimation of depth information in three-dimensional integral imaging", Proc. SPIE 9117, Three-Dimensional Imaging, Visualization, and Display 2014, 911714 (5 June 2014);

Back to Top