boofcv.alg.sfm.d3

Class VisOdomDualTrackPnP<T extends boofcv.struct.image.ImageBase<T>,Desc extends boofcv.struct.feature.TupleDesc>

java.lang.Object

boofcv.alg.sfm.d3.VisOdomDualTrackPnP<T,Desc>

public class VisOdomDualTrackPnP<T extends boofcv.struct.image.ImageBase<T>,Desc extends boofcv.struct.feature.TupleDesc>
extends java.lang.Object

Stereo visual odometry algorithm which relies on tracking features independently in the left and right images and then matching those tracks together. The idea behind this tracker is that the expensive task of association features between left and right cameras only needs to be done once eat time a track is spawned. Triangulation is used to estimate each feature's 3D location. Motion is estimated robustly using a RANSAC type algorithm provided by the user which internally uses PnP type algorithm. Estimated motion is relative to left camera.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	VisOdomDualTrackPnP.LeftTrackInfo
static class	VisOdomDualTrackPnP.RightTrackInfo

Constructor Summary

Constructors

Constructor and Description

VisOdomDualTrackPnP(int thresholdAdd, int thresholdRetire, double epilolarTol, boofcv.abst.feature.tracker.PointTracker<T> trackerLeft, boofcv.abst.feature.tracker.PointTracker<T> trackerRight, boofcv.abst.feature.describe.DescribeRegionPoint<T,Desc> describe, boofcv.abst.feature.associate.AssociateDescription2D<Desc> assocL2R, boofcv.abst.geo.Triangulate2ViewsMetric triangulate, org.ddogleg.fitting.modelset.ModelMatcher<georegression.struct.se.Se3_F64,Stereo2D3D> matcher, org.ddogleg.fitting.modelset.ModelFitter<georegression.struct.se.Se3_F64,Stereo2D3D> modelRefiner) Specifies internal algorithms and parameters

Method Summary

Modifier and Type	Method and Description
java.util.List<boofcv.abst.feature.tracker.PointTrack>	getCandidates() Returns a list of active tracks that passed geometric constraints
georegression.struct.se.Se3_F64	getCurrToWorld()
double	getDescribeRadius()
org.ddogleg.fitting.modelset.ModelMatcher<georegression.struct.se.Se3_F64,Stereo2D3D>	getMatcher()
int	getTick()
boolean	isFault() If there are no candidates then a fault happened.
boolean	process(T left, T right) Updates motion estimate using the stereo pair.
void	reset() Resets the algorithm into its original state
void	setCalibration(boofcv.struct.calib.StereoParameters param)
void	setDescribeRadius(double describeRadius)

Methods inherited from class java.lang.Object

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

Constructor Detail

VisOdomDualTrackPnP

public VisOdomDualTrackPnP(int thresholdAdd,
                           int thresholdRetire,
                           double epilolarTol,
                           boofcv.abst.feature.tracker.PointTracker<T> trackerLeft,
                           boofcv.abst.feature.tracker.PointTracker<T> trackerRight,
                           boofcv.abst.feature.describe.DescribeRegionPoint<T,Desc> describe,
                           boofcv.abst.feature.associate.AssociateDescription2D<Desc> assocL2R,
                           boofcv.abst.geo.Triangulate2ViewsMetric triangulate,
                           org.ddogleg.fitting.modelset.ModelMatcher<georegression.struct.se.Se3_F64,Stereo2D3D> matcher,
                           org.ddogleg.fitting.modelset.ModelFitter<georegression.struct.se.Se3_F64,Stereo2D3D> modelRefiner)

Specifies internal algorithms and parameters

Parameters:

thresholdAdd - When the number of inliers is below this number new features are detected

thresholdRetire - When a feature has not been in the inlier list for this many ticks it is dropped

epilolarTol - Tolerance in pixels for enforcing the epipolar constraint

trackerLeft - Tracker used for left camera

trackerRight - Tracker used for right camera

describe - Describes features in tracks

assocL2R - Assocation for left to right

triangulate - Triangulation for estimating 3D location from stereo pair

matcher - Robust motion model estimation with outlier rejection

modelRefiner - Non-linear refinement of motion model

Method Detail

setCalibration

public void setCalibration(boofcv.struct.calib.StereoParameters param)

reset

public void reset()

Resets the algorithm into its original state

process

public boolean process(T left,
                       T right)

Updates motion estimate using the stereo pair.

Parameters:

left - Image from left camera

right - Image from right camera

Returns:

true if motion estimate was updated and false if not

getCurrToWorld

public georegression.struct.se.Se3_F64 getCurrToWorld()

getTick

public int getTick()

isFault

public boolean isFault()

If there are no candidates then a fault happened.

Returns:

true if fault. false is no fault

getCandidates

public java.util.List<boofcv.abst.feature.tracker.PointTrack> getCandidates()

Returns a list of active tracks that passed geometric constraints

getMatcher

public org.ddogleg.fitting.modelset.ModelMatcher<georegression.struct.se.Se3_F64,Stereo2D3D> getMatcher()

getDescribeRadius

public double getDescribeRadius()

setDescribeRadius

public void setDescribeRadius(double describeRadius)