Three dimensional (3D) object segmentation and tracking can be useful in various computer vision applications, such as: object surveillance for security uses, robot navigation, etc. We present a method for 3D multiple-object tracking using computational integral imaging, based on accurate 3D object segmentation. The method does not employ object detection by motion analysis in a video as conventionally performed (such as background subtraction or block matching). This means that the movement properties do not significantly affect the detection quality. The object detection is performed by analyzing static 3D image data obtained through computational integral imaging With regard to previous works that used integral imaging data in such a scenario, the proposed method performs the 3D tracking of objects without prior information about the objects in the scene, and it is found efficient under severe noise conditions.
Objects in a 3D space can be located and segmented using information obtained by computational integral imaging. This paper implements an approach for objects isolation based on detecting the sharp edges of the focused regions in the reconstructed confocal images. Several edge feature detection methods are employed and examined. Results show that while the ability to detect the correct object depth locations does not depend on the edge detection method, the resulting quality of the detected object features may be significantly affected by the method used.
Recently it was demonstrated that three-dimensional (3D) object recognition and visualization is possible with integral
imaging in photon counting condition or under very low illumination conditions. We present an overview of the
reconstruction techniques, imaging performance and compressive sensing ability of integral imaging in photon starved
Proc. SPIE. 8043, Three-Dimensional Imaging, Visualization, and Display 2011
KEYWORDS: Photon counting, Statistical analysis, 3D acquisition, 3D image reconstruction, Visualization, Image processing, 3D displays, Integral imaging, Expectation maximization algorithms, 3D image processing
Reconstruction of three dimensional (3D) images from photon counting integral images was recently demonstrated by
applying several methods including: maximum likelihood estimation, Bayesian estimation, and statistical estimation
involving truncated Poisson statistics. Here we present simulation results for a new estimation approach implementing
Penalized Maximum Likelihood Expectation Maximization (PMLEM) that better incorporates prior information in the