boofcv.alg.sfm.d3

Class VisOdomQuadPnP<T extends boofcv.struct.image.ImageSingleBand,TD extends boofcv.struct.feature.TupleDesc>

java.lang.Object

boofcv.alg.sfm.d3.VisOdomQuadPnP<T,TD>

public class VisOdomQuadPnP<T extends boofcv.struct.image.ImageSingleBand,TD extends boofcv.struct.feature.TupleDesc>
extends java.lang.Object

Stereo visual odometry algorithm which associates image features across two stereo pairs for a total of four images. Image features are first matched between left and right images while applying epipolar constraints. Then the two more recent sets of stereo images are associated with each other in a left to left and right to right fashion. Features which are consistently matched across all four images are saved in a list. RANSAC is then used to remove false positives and estimate camera motion using a PnP type algorithm. Motion is estimated using PNP algorithms. These require that each image feature as its 3D coordinate estimated. After a feature is associated between a stereo pair its 3D location is also estimated using triangulation. Iterative refinement can then be applied after motion has been estimated. Inside the code each camera is some times referred to by number. 0 = left camera previous frame. 1 = right camera previous frame. 2 = left camera current frame. 3 = right camera current frame. Estimated motion is relative to left camera.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	VisOdomQuadPnP.ImageInfo<TD extends boofcv.struct.feature.TupleDesc> Storage for detected features inside an image
static class	VisOdomQuadPnP.QuadView 3D coordinate of the feature and its observed location in each image
static class	VisOdomQuadPnP.SetMatches Correspondences between images

Constructor Summary

Constructors

Constructor and Description

VisOdomQuadPnP(boofcv.abst.feature.detdesc.DetectDescribeMulti<T,TD> detector, boofcv.abst.feature.associate.AssociateDescription2D<TD> assocSame, boofcv.abst.feature.associate.AssociateDescription2D<TD> assocL2R, boofcv.abst.geo.TriangulateTwoViewsCalibrated 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

Method Summary

Methods

Modifier and Type	Method and Description
georegression.struct.se.Se3_F64	getLeftToWorld()
org.ddogleg.fitting.modelset.ModelMatcher<georegression.struct.se.Se3_F64,Stereo2D3D>	getMatcher()
org.ddogleg.struct.FastQueue<VisOdomQuadPnP.QuadView>	getQuadViews()
boolean	process(T left, T right) Estimates camera egomotion from the stereo pair
void	reset() Resets the algorithm into its original state
void	setCalibration(boofcv.struct.calib.StereoParameters param)

Methods inherited from class java.lang.Object

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

Constructor Detail

VisOdomQuadPnP

public VisOdomQuadPnP(boofcv.abst.feature.detdesc.DetectDescribeMulti<T,TD> detector,
              boofcv.abst.feature.associate.AssociateDescription2D<TD> assocSame,
              boofcv.abst.feature.associate.AssociateDescription2D<TD> assocL2R,
              boofcv.abst.geo.TriangulateTwoViewsCalibrated 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

Parameters:

detector - Estimates image features

assocSame - Association algorithm used for left to left and right to right

assocL2R - Assocation algorithm used for left to right

triangulate - Used to estimate 3D location of a feature using stereo correspondence

matcher - Robust model estimation. Often RANSAC

modelRefiner - Non-linear refinement of motion estimation

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)

Estimates camera egomotion from the stereo pair

Parameters:

left - Image from left camera

right - Image from right camera

Returns:

true if motion was estimated and false if not

getMatcher

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

getQuadViews

public org.ddogleg.struct.FastQueue<VisOdomQuadPnP.QuadView> getQuadViews()

getLeftToWorld

public georegression.struct.se.Se3_F64 getLeftToWorld()