Cameras parameters initialization of 3D kinematic systems

3D information of the scene can be extracted from images acquired by cameras. Before the actual reconstruction camera calibration has to be done. Reconstruction accuracy is highly dictated by the calibration. Two typical demands, which are not easily simultaneously satisfied, are: calibration has to be done in fast and convenient manner and yet assure high degree of reconstruction accuracy. Computational part of calibration usually includes camera parameters initialization and refinement based on initial set of values. The goodness of initial set greatly affects refinement procedure in terms of convergence speed and ultimately reconstruction accuracy. This work proposes new calibration method for 3D kinematic systems. It shortens commonly used calibration procedure, gives better initial parameter values for refinement procedure which in turn is supposed to assure faster and safer convergence of iterative minimization algorithm. Additionally, it will be shown that even without parameter refinement proposed method gives more accurate 3D reconstruction output.