Implementation of an alternative to the POSIT algorithm. More...
#include <ObjectPose.h>
Public Member Functions | |
CObjectPose (const Vec3d *pObjectPoints, int nPoints) | |
bool | EstimatePose (const Vec2d *pImagePoints, Mat3d &rotationMatrix, Vec3d &translationVector, const CCalibration *pCalibration, int nMaxIterations=350) |
~CObjectPose () | |
Private Member Functions | |
void | AbsKernel (Mat3d &R, Vec3d &T, float &error) |
Vec3d | tEstimate (Mat3d &R) |
Private Attributes | |
int | m_nPoints |
Mat3d * | m_pF |
Vec3d * | m_pP |
Vec3d * | m_pQ |
Mat3d | m_tFactor |
Vec3d | m_tPbar |
Implementation of an alternative to the POSIT algorithm.
This class implements the approach presented in (C.-P. Lu, G. D. Hager and E. Mjolsness, "Fast and Globally Convergent Pose Estimation from Video Images", 2000).
This algorithm also succeeds in the case of planar points sets (objects).
Definition at line 74 of file ObjectPose.h.
CObjectPose::CObjectPose | ( | const Vec3d * | pObjectPoints, |
int | nPoints | ||
) |
Definition at line 84 of file ObjectPose.cpp.
CObjectPose::~CObjectPose | ( | ) |
Definition at line 109 of file ObjectPose.cpp.
Definition at line 197 of file ObjectPose.cpp.
bool CObjectPose::EstimatePose | ( | const Vec2d * | pImagePoints, |
Mat3d & | rotationMatrix, | ||
Vec3d & | translationVector, | ||
const CCalibration * | pCalibration, | ||
int | nMaxIterations = 350 |
||
) |
Definition at line 121 of file ObjectPose.cpp.
Definition at line 307 of file ObjectPose.cpp.
|
private |
Definition at line 96 of file ObjectPose.h.
|
private |
Definition at line 101 of file ObjectPose.h.
|
private |
Definition at line 95 of file ObjectPose.h.
|
private |
Definition at line 99 of file ObjectPose.h.
|
private |
Definition at line 102 of file ObjectPose.h.
|
private |
Definition at line 97 of file ObjectPose.h.