Fig 1.
An incident ray passing through an imaging system which is absorbed by detector elements (pixel).
Fig 2.
Geometrization of the IRT.
Fig 3.
A mapping between reference plane and image plane.
The intersection points of dotted lines in plane ABCD corresponds to the ones in plane abcd.
Fig 4.
The perspective rays used for object pose.
Fig 5.
An imaging model of the perspective rays.
It includes a perspective projection and a scaled orthographic projection.
Fig 6.
Captured images at six positions.
Table 1.
Camera parameters.
Fig 7.
Position distribution of the calibration points.
The “●” represents the standard two-dimensional coordinate of the calibration points while the “+” represents the calculated two-dimensional coordinate of them.
Fig 8.
Experiment devices.
Fig 9.
A sample image used for pose estimation.
Fig 10.
Pose estimation error distribution.
Table 2.
The RMSE of the PnP algorithms.