boofcv.alg.geo.bundle

Class CalibratedPoseAndPoint

java.lang.Object

boofcv.alg.geo.bundle.CalibratedPoseAndPoint

public class CalibratedPoseAndPoint
extends java.lang.Object

Expanded model for fast computations used by bundle adjustment with calibrated cameras. Designed to minimize unnecessary creating and destroying memory. Each view can be marked as known or not. If a view is known then initial pose is fixed and will not be optimized. The list of points is a list of all points which have been observed across all views. Both views and points are referenced by their index. Before the class is used you must call configure(int, int) to specif the number of views and points.

Constructor Summary

Constructors

Constructor and Description
CalibratedPoseAndPoint()

Method Summary

Modifier and Type	Method and Description
void	configure(int numViews, int numPoints) Specifies the number of views and 3D points being estimated
boolean[]	getKnownArray() An array that indicates which views are known and which views are not
int	getNumPoints() The number of points (or features) whose location is being optimized.
int	getNumUnknownViews() Returns the number of camera views which do not have a 'known' or fixed pose.
int	getNumViews() The number of camera views which observed the points/features
georegression.struct.point.Point3D_F64	getPoint(int index) Returns the location of a specific point/feature
georegression.struct.se.Se3_F64	getWorldToCamera(int view) Transform from world to camera view
boolean	isViewKnown(int view) Used to see if a particular view is marked as known or not
void	setViewKnown(int view, boolean known) Specify if a view is assumed to have a known view or not.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

CalibratedPoseAndPoint

public CalibratedPoseAndPoint()

Method Detail

configure

public void configure(int numViews,
                      int numPoints)

Specifies the number of views and 3D points being estimated

Parameters:

numViews - Number of camera views observing the points.

numPoints - Number of points observed

setViewKnown

public void setViewKnown(int view,
                         boolean known)

Specify if a view is assumed to have a known view or not.

Parameters:

view - Index of the view

known - true of known or false if not.

isViewKnown

public boolean isViewKnown(int view)

Used to see if a particular view is marked as known or not

Parameters:

view - The view's index

Returns:

if true then the view's pose is assumed to be known and is not optimized

getWorldToCamera

public georegression.struct.se.Se3_F64 getWorldToCamera(int view)

Transform from world to camera view

Parameters:

view - The view's index

Returns:

The transform

getPoint

public georegression.struct.point.Point3D_F64 getPoint(int index)

Returns the location of a specific point/feature

Parameters:

index - Index of the point

Returns:

The point's location

getNumPoints

public int getNumPoints()

The number of points (or features) whose location is being optimized.

Returns:

number of points

getNumViews

public int getNumViews()

The number of camera views which observed the points/features

Returns:

Number of views

getNumUnknownViews

public int getNumUnknownViews()

Returns the number of camera views which do not have a 'known' or fixed pose.

Returns:

Number of views whose pose is to be optimized.

getKnownArray

public boolean[] getKnownArray()

An array that indicates which views are known and which views are not