com.irurueta.ar.sfm.BaseAbsoluteOrientationSlamSparseReconstructor Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of irurueta-ar Show documentation
Show all versions of irurueta-ar Show documentation
Augmented Reality and 3D reconstruction library
/*
* Copyright (C) 2017 Alberto Irurueta Carro ([email protected])
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.irurueta.ar.sfm;
import com.irurueta.ar.slam.AbsoluteOrientationBaseSlamEstimator;
import com.irurueta.ar.slam.BaseCalibrationData;
import com.irurueta.geometry.MetricTransformation3D;
import com.irurueta.geometry.PinholeCamera;
import com.irurueta.geometry.Point3D;
import com.irurueta.geometry.Rotation3D;
import java.util.ArrayList;
import java.util.List;
/**
* Base class in charge of estimating cameras and 3D reconstructed points from sparse
* image point correspondences in multiple views and also in charge of estimating overall
* scene scale and absolute orientation by means of SLAM (Simultaneous Location And Mapping)
* using data obtained from sensors like accelerometers or gyroscopes.
* NOTE: absolute orientation slam estimators are not very accurate during estimation of
* the orientation state, for that reason we take into account the initial orientation.
*
* @param type defining calibration data.
* @param type of configuration.
* @param type of re-constructor.
* @param type of listener.
* @param type of SLAM estimator.
*/
public abstract class BaseAbsoluteOrientationSlamSparseReconstructor<
D extends BaseCalibrationData,
C extends BaseSlamSparseReconstructorConfiguration,
R extends BaseSlamSparseReconstructor,
L extends BaseSlamSparseReconstructorListener,
S extends AbsoluteOrientationBaseSlamEstimator> extends
BaseSlamSparseReconstructor {
/**
* First sample of orientation received.
*/
private Rotation3D mFirstOrientation;
/**
* Inverse of first orientation.
*/
private Rotation3D mInvFirstOrientation;
/**
* Constructor.
*
* @param configuration configuration for this re-constructor.
* @param listener listener in charge of handling events.
* @throws NullPointerException if listener or configuration is not
* provided.
*/
protected BaseAbsoluteOrientationSlamSparseReconstructor(
final C configuration, final L listener) {
super(configuration, listener);
}
/**
* Provides a new orientation sample to update SLAM estimator.
* If re-constructor is not running, calling this method has no effect.
*
* @param timestamp timestamp of accelerometer sample since epoch time and
* expressed in nanoseconds.
* @param orientation new orientation.
*/
public void updateOrientationSample(final long timestamp,
final Rotation3D orientation) {
if (mSlamEstimator != null) {
mSlamEstimator.updateOrientationSample(timestamp, orientation);
}
if (mFirstOrientation == null) {
// make a copy of orientation
mFirstOrientation = orientation.toQuaternion();
mInvFirstOrientation = mFirstOrientation.inverseRotationAndReturnNew();
}
}
/**
* Updates scene scale and orientation using SLAM data.
*
* @param isInitialPairOfViews true if initial pair of views is being processed, false otherwise.
* @return true if scale was successfully updated, false otherwise.
*/
@SuppressWarnings("DuplicatedCode")
protected boolean updateScaleAndOrientation(final boolean isInitialPairOfViews) {
try {
final PinholeCamera metricCamera1 = mPreviousMetricEstimatedCamera.getCamera();
final PinholeCamera metricCamera2 = mCurrentMetricEstimatedCamera.getCamera();
double slamPosX;
double slamPosY;
double slamPosZ;
double scale;
if (isInitialPairOfViews) {
// obtain baseline (camera separation from slam estimator data
slamPosX = mSlamEstimator.getStatePositionX();
slamPosY = mSlamEstimator.getStatePositionY();
slamPosZ = mSlamEstimator.getStatePositionZ();
mSlamPosition.setInhomogeneousCoordinates(slamPosX, slamPosY, slamPosZ);
metricCamera1.decompose(false, true);
metricCamera2.decompose(false, true);
final Point3D center1 = metricCamera1.getCameraCenter();
final Point3D center2 = metricCamera2.getCameraCenter();
final double baseline = center1.distanceTo(mSlamPosition);
final double estimatedBaseline = center1.distanceTo(center2);
scale = mCurrentScale = baseline / estimatedBaseline;
} else {
scale = mCurrentScale;
}
// R1' = R1*Rdiff
// Rdiff = R1^T*R1'
// where R1' is the desired orientation (obtained by sampling a
// sensor)
// and R1 is always the identity for the 1st camera.
// Hence R1' = Rdiff
// t1' is the desired translation which is zero for the 1st
// camera.
// We want: P1' = K*[R1' t1'] = K*[R1' 0]
// And we have P1 = K[I 0]
// We need a transformation T so that:
// P1' = P1*T^-1 = K[I 0][R1' 0]
// [0 1]
// Hence: T^-1 = [R1' 0]
// [0 1]
// or T = [R1'^T 0]
// [0 1]
// because we are also applying a transformation of scale s,
// the combination of both transformations is
// T = [s*R1'^T 0]
// [0 1]
final MetricTransformation3D scaleAndOrientationTransformation =
new MetricTransformation3D(scale);
scaleAndOrientationTransformation.setRotation(mInvFirstOrientation);
// update scale of cameras
final PinholeCamera euclideanCamera1 = scaleAndOrientationTransformation.transformAndReturnNew(
metricCamera1);
final PinholeCamera euclideanCamera2 = scaleAndOrientationTransformation.transformAndReturnNew(
metricCamera2);
euclideanCamera2.decompose(false, true);
mSlamEstimator.correctWithPositionMeasure(euclideanCamera2.getCameraCenter(),
mConfiguration.getCameraPositionCovariance());
if (!isInitialPairOfViews) {
slamPosX = mSlamEstimator.getStatePositionX();
slamPosY = mSlamEstimator.getStatePositionY();
slamPosZ = mSlamEstimator.getStatePositionZ();
mSlamPosition.setInhomogeneousCoordinates(slamPosX, slamPosY, slamPosZ);
// adjust scale of current camera
final Point3D euclideanCenter2 = euclideanCamera2.getCameraCenter();
final double euclideanPosX = euclideanCenter2.getInhomX();
final double euclideanPosY = euclideanCenter2.getInhomY();
final double euclideanPosZ = euclideanCenter2.getInhomZ();
final double scaleVariationX = euclideanPosX / slamPosX;
final double scaleVariationY = euclideanPosY / slamPosY;
final double scaleVariationZ = euclideanPosZ / slamPosZ;
final double scaleVariation = (scaleVariationX + scaleVariationY + scaleVariationZ) / 3.0;
scale *= scaleVariation;
mCurrentScale = scale;
scaleAndOrientationTransformation.setScale(mCurrentScale);
// update camera
scaleAndOrientationTransformation.transform(metricCamera2, euclideanCamera2);
}
final double sqrScale = scale * scale;
mPreviousEuclideanEstimatedCamera = new EstimatedCamera();
mPreviousEuclideanEstimatedCamera.setCamera(euclideanCamera1);
mPreviousEuclideanEstimatedCamera.setViewId(mPreviousMetricEstimatedCamera.getViewId());
mPreviousEuclideanEstimatedCamera.setQualityScore(mPreviousMetricEstimatedCamera.getQualityScore());
if (mPreviousMetricEstimatedCamera.getCovariance() != null) {
mPreviousEuclideanEstimatedCamera.setCovariance(
mPreviousMetricEstimatedCamera.getCovariance().multiplyByScalarAndReturnNew(sqrScale));
}
mCurrentEuclideanEstimatedCamera = new EstimatedCamera();
mCurrentEuclideanEstimatedCamera.setCamera(euclideanCamera2);
mCurrentEuclideanEstimatedCamera.setViewId(mCurrentMetricEstimatedCamera.getViewId());
mCurrentEuclideanEstimatedCamera.setQualityScore(mCurrentMetricEstimatedCamera.getQualityScore());
if (mCurrentMetricEstimatedCamera.getCovariance() != null) {
mCurrentEuclideanEstimatedCamera.setCovariance(
mCurrentMetricEstimatedCamera.getCovariance().multiplyByScalarAndReturnNew(sqrScale));
}
// update scale of reconstructed points
final int numPoints = mActiveMetricReconstructedPoints.size();
final List metricReconstructedPoints3D = new ArrayList<>();
for (final ReconstructedPoint3D reconstructedPoint : mActiveMetricReconstructedPoints) {
metricReconstructedPoints3D.add(reconstructedPoint.getPoint());
}
final List euclideanReconstructedPoints3D =
scaleAndOrientationTransformation.transformPointsAndReturnNew(
metricReconstructedPoints3D);
// set scaled points into result
mActiveEuclideanReconstructedPoints = new ArrayList<>();
ReconstructedPoint3D euclideanPoint;
ReconstructedPoint3D metricPoint;
for (int i = 0; i < numPoints; i++) {
metricPoint = mActiveMetricReconstructedPoints.get(i);
euclideanPoint = new ReconstructedPoint3D();
euclideanPoint.setId(metricPoint.getId());
euclideanPoint.setPoint(euclideanReconstructedPoints3D.get(i));
euclideanPoint.setInlier(metricPoint.isInlier());
euclideanPoint.setQualityScore(metricPoint.getQualityScore());
if (metricPoint.getCovariance() != null) {
euclideanPoint.setCovariance(metricPoint.getCovariance().multiplyByScalarAndReturnNew(sqrScale));
}
euclideanPoint.setColorData(metricPoint.getColorData());
mActiveEuclideanReconstructedPoints.add(euclideanPoint);
}
return true;
} catch (final Exception e) {
mFailed = true;
//noinspection unchecked
mListener.onFail((R) this);
return false;
}
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy