Since a large number of vision applications rely on the mapping between 3D scenes and their corresponding 2D camera images, an important practical consideration for researchers is, what are the major determinants of camera calibration accuracy and what accuracy can be achieved within the practical limits of their environments. In response, we present a thorough study investigating the effects of training data quantity, measurement error, pixel coordinate noise, and the choice of camera model, on camera calibration results. Through this effort, we seek to determine whether expensive, elaborate setups are necessary, or indeed, beneficial, to camera calibration, and whether a high complexity camera model leads to improved accuracy. The results are first provided for a simulated camera system and then verified through carefully controlled experiments using real-world measurements.
Requirements for Camera Calibration: Must Accuracy Come with a High Price?
2005-01-01
245513 byte
Conference paper
Electronic Resource
English
Secretaries must come to grips
Online Contents | 1994
Accuracy requirements in off-line array calibration
IEEE | 1997
|Accuracy Requirements in Off-Line Array Calibration
Online Contents | 1997
|Editorials - Survey shows: Customers must still come first
Online Contents | 2006