Package boofcv.alg.geo.trifocal
Class TrifocalLinearPoint7
java.lang.Object
boofcv.alg.geo.trifocal.TrifocalLinearPoint7
- Direct Known Subclasses:
TrifocalAlgebraicPoint7
Estimates the TrifocalTensor
using a linear algorithm from 7 or more image points correspondences
from three views, see page 394 of [1] for details. After an initial linear solution has been computed
it is improved upon by applying geometric constraints. Note that the solution will not be optimal in a geometric
or algebraic sense, but can be used as an initial estimate for refinement algorithms.
References:
- R. Hartley, and A. Zisserman, "Multiple View Geometry in Computer Vision", 2nd Ed, Cambridge 2003
- See Also:
-
Field Summary
Modifier and TypeFieldDescriptionprotected DMatrixRMaj
protected Point3D_F64
protected Point3D_F64
protected EnforceTrifocalGeometry
protected TrifocalExtractGeometries
protected NormalizationPoint2D
protected NormalizationPoint2D
protected NormalizationPoint2D
protected Point2D_F64
protected Point2D_F64
protected Point2D_F64
protected TrifocalTensor
protected SingularValueDecomposition_F64<DMatrixRMaj>
protected DMatrixRMaj
-
Constructor Summary
-
Method Summary
Modifier and TypeMethodDescriptionprotected void
createLinearSystem
(List<AssociatedTriple> observations) Constructs the linear matrix that describes from the 3-point constraint with linear dependent rows removedboolean
process
(List<AssociatedTriple> observations, TrifocalTensor solution) Estimates the trifocal tensor given the set of observationsprotected void
removeNormalization
(TrifocalTensor solution) Translates the trifocal tensor back into regular coordinate systemprotected boolean
Computes the null space of the linear system to find the trifocal tensor
-
Field Details
-
solutionN
-
N1
-
N2
-
N3
-
A
-
svdNull
-
vectorizedSolution
-
p1_norm
-
p2_norm
-
p3_norm
-
enforce
-
extractEpipoles
-
e2
-
e3
-
-
Constructor Details
-
TrifocalLinearPoint7
public TrifocalLinearPoint7()
-
-
Method Details
-
process
Estimates the trifocal tensor given the set of observations- Parameters:
observations
- Set of observationssolution
- Output: Where the solution is written to- Returns:
- true if successful and false if it fails
-
createLinearSystem
Constructs the linear matrix that describes from the 3-point constraint with linear dependent rows removed -
solveLinearSystem
protected boolean solveLinearSystem()Computes the null space of the linear system to find the trifocal tensor -
removeNormalization
Translates the trifocal tensor back into regular coordinate system
-