Calibrating a one-dimensional coded light 3D acquisition system
A camera intrinsic calibration may be performed using an object geometry. An intrinsic camera matrix may then be recovered. A homography is fit between object and camera coordinate systems. View transformations are finally recovered.
1. A method comprising:
performing a system calibration for a one-dimensional coded light three-dimensional acquisition system by:
using a coarse calibration to estimate system parameters;
minimizing geometric reconstruction error including iterating until convergence of the following refinement sequence:
fixing all parameters except a lens distortion model to determine camera distortion;
fixing all parameters except a 2×4 projection matrix P P to determine a projector matrix P P ;
fixing reconstructed points to determine an error function for an n-th view;
fixing all parameters except parameters for a known planar calibration object and minimizing for a final bundle adjustment;
refining R P and t P using the refinement sequence as the initial value and minimizing geometric error, where R P and t P are extrinsic rotation and translation parameters between the camera and projector coordinate system; and
recovering K P , R P and t P from the projector matrix P P , where K P is a 2×3 intrinsic matrix.
2. The method of claim 1 including using system parameters and view transformation as optimization variables.
3. The method of claim 2 including fixing all parameters except an intrinsic camera matrix to determine camera distortion.
4. One or more non-transitory computer readable media storing instructions to perform a sequence comprising:
performing a system calibration for a one-dimensional coded light three-dimensional acquisition system by:
using a coarse calibration to estimate system parameters;
minimizing geometric reconstruction error including iterating until convergence of the following refinement sequence:
fixing all parameters except a lens distortion model to determine camera distortion;
fixing all parameters except a 2×4 projection matrix P P to determine a projector matrix P P ;
fixing reconstructed points to determine an error function for an n-th view;
fixing all parameters except parameters for a known planar calibration object and minimizing for a final bundle adjustment;
refining R P and t P using the refinement sequence as the initial value and minimizing geometric error, where R P and t P are extrinsic rotation and translation parameters between the camera and projector coordinate system; and
recovering K P , R P and t P from the projector matrix P P , where K P is a 2×3 intrinsic matrix.
5. The media of claim 4 , further storing instructions to perform a sequence including using system parameters and view transformation as optimization variables.
6. The media of claim 5 , further storing instructions to perform a sequence including fixing all parameters except an intrinsic camera matrix to determine camera distortion.
7. An apparatus comprising:
a processor to perform a system calibration for a one-dimensional coded light three-dimensional acquisition system by using a coarse calibration to estimate system parameters, minimize geometric reconstruction error including iterating until convergence of the following refinement sequence:
fix all parameters except a lens distortion model to determine camera distortion;
fix all parameters except a 2×4 projection matrix P P to determine a projector matrix P P ;
fix reconstructed points to determine an error function for an n-th view;
fix all parameters except parameters for a known planar calibration object and minimizing for a final bundle adjustment;
refine R P and t P using the refinement sequence as the initial value and minimizing geometric error where R P and t P are extrinsic rotation and translation parameters between the camera and projector coordinate system, and recover K P , R P and t P from the projector matrix P P , where K P is a 2×3 intrinsic matrix.
8. The apparatus of claim 7 , said processor to use system parameters and view transformation as optimization variables.
9. The apparatus of claim 8 , said processor to fix all parameters except an intrinsic camera matrix to determine camera distortion.