Camera calibration with three noncollinear points under special motions

Area of Science:

• Computer Vision
• Robotics
• Computational Geometry

Background:

• Plane-based (2-D) camera calibration is flexible but requires at least four points per view.
• Existing 1-D calibration methods use three collinear points, a restrictive setup.
• A need exists for camera calibration methods with fewer constraints on calibration objects.

Purpose of the Study:

• To propose a new camera calibration algorithm using three noncollinear points.
• To investigate the feasibility and advantages of using minimal point configurations for calibration.
• To bridge the gap between 1-D and 2-D calibration techniques.

Main Methods:

• Development of a novel camera calibration algorithm utilizing three noncollinear points.
• Theoretical validation of the proposed method.
• Experimental verification using simulated and real image data.

Main Results:

• The proposed algorithm successfully calibrates cameras using only three noncollinear points.
• Experiments demonstrate the theoretical correctness and numerical robustness of the method.
• The three-point noncollinear configuration is shown to be a versatile intermediate between 1-D and 2-D calibration.

Conclusions:

• A new, efficient camera calibration algorithm has been developed.
• The method offers a practical solution for scenarios where traditional four-point methods are challenging.
• This work advances camera calibration by introducing a novel approach with reduced object constraints.