Extracting 3D from 2D: Selection Basis for Camera Calibration

M.-C. Villa-Uriol, G. Chaudhary, F. Kuester, T. Hutchinson, and N. Bagherzadeh (USA)


Camera calibration, computer vision, scientific visualiza tion.


The recovery of three-dimensional information from a set of two-dimensional images requires the use of an accurate, and fast calibration technique. Although diverse techniques have been developed for single and multiple camera con figurations, selection of the most appropriate task-specific technique is a non-trivial exercise. Therefore, in this pa per, we present a review of calibration techniques, their re spective application domains and associated boundary con ditions. Boundary conditions of particular importance in clude, desired capture volume, available time budget, re quired accuracy, number of cameras and type of feature de tection. Commonly used calibration tools are discussed, including their specific features and target applications. A methodology to determine the suitability of a calibration tool for a specific setup and requirements using a syntheti cally generated environment, is presented. Finally, a multi camera calibration tool is evaluated for its robustness in de termining important calibration information under a num ber of envisioned camera configurations. The goal of this study is to aid the reader in the selection of the most suit able calibration techniques and tools appropriate for the task at hand.

Important Links:

Go Back