TemplateTrackerWarpHomography¶

class TemplateTrackerWarpHomography(self)¶

Bases: TemplateTrackerWarp

This class consider the homography warping model \(M\) with parameters \(p=(h_1, h_2, h_3, h_4, h_5, h_6, h_7, h_8)\) such as

\[\begin{split}M(p) = \left[ \begin{array}{ccc} h_1 + 1 & h_4 & h_7 \\h_2 & h_5 + 1 & h_8 \\h_3 & h_6 & 1 \end{array} \right] \end{split}\]

We recall that u axis is the image horizontal axis, and v axis is the image vertical axis. A point (u,v) with coordinates (0,0) is located in the top left image corner.

Construct an homography model with 8 parameters initialized to zero.

Methods

`__init__`	Construct an homography model with 8 parameters initialized to zero.
`computeDenom`	Compute the projection denominator (Z) used in x = X/Z and y = Y/Z.
`dWarp`	Compute the derivative matrix of the warping function at point \(X=(u,v)\) according to the model parameters:
`getHomography`	return: Corresponding homography.
`getParam`	Overloaded function.
`getParamInverse`	Compute inverse of the warping transformation.
`getParamPyramidDown`	Get the parameters of the warping function one level down where image size is divided by two along the lines and the columns.
`getParamPyramidUp`	Get the parameters of the warping function one level up where image size is multiplied by two along the lines and the columns.
`isESMcompatible`	return: false. Homography model is not compatible with ESM.
`pRondp`	Compute the transformation resulting from the composition of two other transformations.
`warpX`	Overloaded function.
`warpXInv`	Warp a point X1 with the inverse transformation \(M\) .

Inherited Methods

`warpTriangle`	Warp a triangle and store the result in a new zone.
`setNbParam`	Set the number of parameters of the warping function.
`getDistanceBetweenZoneAndWarpedZone`	Compute the distance between a zone and its associated warped zone.
`getNbParam`	Get the number of parameters of the warping function.
`warpZone`	Warp a zone and store the result in a new zone.

Operators

`__doc__`
`__init__`	Construct an homography model with 8 parameters initialized to zero.
`__module__`

Attributes

`__annotations__`

__init__(self)¶: Construct an homography model with 8 parameters initialized to zero.

computeDenom(self, X: visp._visp.core.ColVector, p: visp._visp.core.ColVector) → None¶

Compute the projection denominator (Z) used in x = X/Z and y = Y/Z.

Note

See warpX(const vpColVector &, vpColVector &, const vpColVector &)

Note

See warpX(const int &, const int &, double &, double &, const vpColVector &)

Note

See dWarp() , dWarpCompo()

Parameters:

X: visp._visp.core.ColVector¶: Point with coordinates (u, v) to consider.
p: visp._visp.core.ColVector¶: 8-dim vector that contains the current parameters of the warping function.

dWarp(self, arg0: visp._visp.core.ColVector, X: visp._visp.core.ColVector, arg2: visp._visp.core.ColVector, dW: visp._visp.core.Matrix) → None¶

Compute the derivative matrix of the warping function at point \(X=(u,v)\) according to the model parameters:

\[\frac{\partial M}{\partial p}(X, p) \]

Note

See computeDenom()

Parameters:

X: visp._visp.core.ColVector¶: 2-dim vector corresponding to the coordinates \((u_1, v_1)\) of the point to consider in the derivative computation.

getDistanceBetweenZoneAndWarpedZone(self, Z: visp._visp.tt.TemplateTrackerZone, p: visp._visp.core.ColVector) → float¶

Compute the distance between a zone and its associated warped zone.

Parameters:

Z: visp._visp.tt.TemplateTrackerZone¶: Zone to consider.
p: visp._visp.core.ColVector¶: Parameters of the warping function.

getHomography(self, ParamM: visp._visp.core.ColVector) → visp._visp.vision.Homography¶

Returns:: Corresponding homography.

getNbParam(self) → int¶

Get the number of parameters of the warping function.

Returns:: Number of parameters.

getParam(*args, **kwargs)¶

Overloaded function.

getParam(self: visp._visp.tt.TemplateTrackerWarpHomography, H: visp._visp.vision.Homography, p: visp._visp.core.ColVector) -> None

Compute the parameters corresponding to an homography.

Parameters:

H: Homography
p: 8-dim vector corresponding to the homography.

getParam(self: visp._visp.tt.TemplateTrackerWarpHomography, H: visp._visp.core.Matrix, p: visp._visp.core.ColVector) -> None

Compute the parameters corresponding to an homography as a 3-by-3 matrix.

Parameters:

H: 3-by-3 matrix corresponding to an homography
p: 8-dim vector corresponding to the homography.

getParamInverse(self, p: visp._visp.core.ColVector, p_inv: visp._visp.core.ColVector) → None¶

Compute inverse of the warping transformation.

Parameters:

p: visp._visp.core.ColVector¶: 8-dim vector that contains the parameters corresponding to the transformation to inverse.
p_inv: visp._visp.core.ColVector¶: 8-dim vector that contains the parameters of the inverse transformation \({M(p)}^{-1}\) .

getParamPyramidDown(self, p: visp._visp.core.ColVector, p_down: visp._visp.core.ColVector) → None¶

Get the parameters of the warping function one level down where image size is divided by two along the lines and the columns.

Parameters:

p: visp._visp.core.ColVector¶: 8-dim vector that contains the current parameters of the warping function.
p_down: visp._visp.core.ColVector¶: 8-dim vector that contains the resulting parameters one level down.

getParamPyramidUp(self, p: visp._visp.core.ColVector, p_up: visp._visp.core.ColVector) → None¶

Get the parameters of the warping function one level up where image size is multiplied by two along the lines and the columns.

Parameters:

p: visp._visp.core.ColVector¶: 8-dim vector that contains the current parameters of the warping function.
p_up: visp._visp.core.ColVector¶: 8-dim vector that contains the resulting parameters one level up.

isESMcompatible(self) → bool¶

Returns:: false. Homography model is not compatible with ESM.

pRondp(self, p1: visp._visp.core.ColVector, p2: visp._visp.core.ColVector, p12: visp._visp.core.ColVector) → None¶

Compute the transformation resulting from the composition of two other transformations.

Parameters:

p1: visp._visp.core.ColVector¶: 8-dim vector that contains the parameters corresponding to first transformation.
p2: visp._visp.core.ColVector¶: 8-dim vector that contains the parameters corresponding to second transformation.
p12: visp._visp.core.ColVector¶: 8-dim vector that contains the resulting transformation \(p_{12} = p_1 \circ p_2\) .

setNbParam(self, nb: int) → None¶

Set the number of parameters of the warping function.

Parameters:

nb: int¶: New number of parameters.

warpTriangle(self: visp._visp.tt.TemplateTrackerWarp, in: visp._visp.tt.TemplateTrackerTriangle, p: visp._visp.core.ColVector, out: visp._visp.tt.TemplateTrackerTriangle) → None¶

Warp a triangle and store the result in a new zone.

Parameters:

in: Triangle to warp.
p: Parameters of the warping function. These parameters are estimated by the template tracker and returned using vpTemplateTracker::getp() .
out: Resulting triangle.

warpX(*args, **kwargs)¶

Overloaded function.

warpX(self: visp._visp.tt.TemplateTrackerWarpHomography, X1: visp._visp.core.ColVector, X2: visp._visp.core.ColVector, p: visp._visp.core.ColVector) -> None

Warp point \(X_1=(u_1,v_1)\) using the transformation model.

\[X_2 = {^2}M_1(p) * X_1\]

Note

See computeDenom()

Parameters:

X1: 2-dim vector corresponding to the coordinates \((u_1, v_1)\) of the point to warp.
X2: 2-dim vector corresponding to the coordinates \((u_2, v_2)\) of the warped point.
p: 8-dim vector that contains the parameters of the transformation.

warpX(self: visp._visp.tt.TemplateTrackerWarpHomography, v1: int, u1: int, v2: float, u2: float, p: visp._visp.core.ColVector) -> tuple[float, float]

Warp point \(X_1=(u_1,v_1)\) using the transformation model with parameters \(p\) .

\[X_2 = {^2}M_1(p) * X_1\]

Parameters:

v1: Coordinate (along the image rows axis) of the point \(X_1=(u_1,v_1)\) to warp.
u1: Coordinate (along the image columns axis) of the point \(X_1=(u_1,v_1)\) to warp.
v2: Coordinate of the warped point \(X_2=(u_2,v_2)\) along the image rows axis.
u2: Coordinate of the warped point \(X_2=(u_2,v_2)\) along the image column axis.
p: 8-dim vector that contains the parameters of the transformation.

Returns:

A tuple containing:

v2: Coordinate of the warped point \(X_2=(u_2,v_2)\) along the image rows axis.
u2: Coordinate of the warped point \(X_2=(u_2,v_2)\) along the image column axis.

warpXInv(self, X1: visp._visp.core.ColVector, X2: visp._visp.core.ColVector, p: visp._visp.core.ColVector) → None¶

Warp a point X1 with the inverse transformation \(M\) .

\[X_2 = {\left({^1}M_2\right)}^{-1} \; X_1\]

Parameters:

X1: visp._visp.core.ColVector¶: 2-dim vector corresponding to the coordinates (u,v) of the point to warp.
X2: visp._visp.core.ColVector¶: 2-dim vector corresponding to the coordinates (u,v) of the warped point.
p: visp._visp.core.ColVector¶: Parameters corresponding to the warping model \({^1}M_2\) .

warpZone(self: visp._visp.tt.TemplateTrackerWarp, in: visp._visp.tt.TemplateTrackerZone, p: visp._visp.core.ColVector, out: visp._visp.tt.TemplateTrackerZone) → None¶

Warp a zone and store the result in a new zone.

Parameters:

in: Zone to warp.
p: Parameters of the warping function. These parameters are estimated by the template tracker and returned using vpTemplateTracker::getp() .
out: Resulting zone.