Package com.irurueta.ar.epipolar.estimators

Class PROMedSFundamentalMatrixRobustEstimator

java.lang.Object
com.irurueta.ar.epipolar.estimators.FundamentalMatrixRobustEstimator
com.irurueta.ar.epipolar.estimators.PROMedSFundamentalMatrixRobustEstimator
public class PROMedSFundamentalMatrixRobustEstimator extends FundamentalMatrixRobustEstimator
Finds the best fundamental matrix for provided collections of matched 2D points using PROMedS algorithm.

  • Field Details

    • DEFAULT_PROMEDS_FUNDAMENTAL_MATRIX_ESTIMATOR_METHOD

      public static final FundamentalMatrixEstimatorMethod DEFAULT_PROMEDS_FUNDAMENTAL_MATRIX_ESTIMATOR_METHOD
      Default non-robust method to estimate a fundamental matrix.

    • DEFAULT_STOP_THRESHOLD

      public static final double DEFAULT_STOP_THRESHOLD
      Default value to be used for stop threshold. Stop threshold can be used to keep the algorithm iterating in case that best estimated threshold using median of residuals is not small enough. Once a solution is found that generates a threshold below this value, the algorithm will stop. The stop threshold can be used to prevent the LMedS algorithm iterating too many times in cases where samples have a very similar accuracy. For instance, in cases where proportion of outliers is very small (close to 0%), and samples are very accurate (i.e. 1e-6), the algorithm would iterate for a long time trying to find the best solution when indeed there is no need to do that if a reasonable threshold has already been reached. Because of this behaviour the stop threshold can be set to a value much lower than the one typically used in RANSAC, and yet the algorithm could still produce even smaller thresholds in estimated results.
      See Also:

    • MIN_STOP_THRESHOLD

      public static final double MIN_STOP_THRESHOLD
      Minimum allowed stop threshold value.
      See Also:

    • stopThreshold

      private double stopThreshold
      Threshold to be used to keep the algorithm iterating in case that best estimated threshold using median of residuals is not small enough. Once a solution is found that generates a threshold below this value, the algorithm will stop. The stop threshold can be used to prevent the LMedS algorithm iterating too many times in cases where samples have a very similar accuracy. For instance, in cases where proportion of outliers is very small (close to 0%), and samples are very accurate (i.e. 1e-6), the algorithm would iterate for a long time trying to find the best solution when indeed there is no need to do that if a reasonable threshold has already been reached. Because of this behaviour the stop threshold can be set to a value much lower than the one typically used in RANSAC, and yet the algorithm could still produce even smaller thresholds in estimated results.

    • qualityScores

      private double[] qualityScores
      Quality scores corresponding to each provided point. The larger the score value the better the quality of the sample.

  • Constructor Details

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixEstimatorMethod fundMatrixEstimatorMethod)
      Constructor.
      Parameters:
      fundMatrixEstimatorMethod - method for non-robust fundamental matrix estimator.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixEstimatorMethod fundMatrixEstimatorMethod, FundamentalMatrixRobustEstimatorListener listener)
      Constructor.
      Parameters:
      fundMatrixEstimatorMethod - method for non-robust fundamental matrix estimator.
      listener - listener to be notified of events such as when estimation starts, ends or its progress significantly changes.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixEstimatorMethod fundMatrixEstimatorMethod, List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints)
      Constructor.
      Parameters:
      fundMatrixEstimatorMethod - method for non-robust fundamental matrix estimator.
      leftPoints - 2D points on left view.
      rightPoints - 2D points on right view.
      Throws:
      IllegalArgumentException - if provided list of points do not have the same length or their length is less than 8 points.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixEstimatorMethod fundMatrixEstimatorMethod, List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints, FundamentalMatrixRobustEstimatorListener listener)
      Constructor.
      Parameters:
      fundMatrixEstimatorMethod - method for non-robust fundamental matrix estimator.
      leftPoints - 2D points on left view.
      rightPoints - 2D points on right view.
      listener - listener to be notified of events such as when estimation starts, ends or its progress significantly changes.
      Throws:
      IllegalArgumentException - if provided list of points do not have the same length or their length is less than 8 points.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixEstimatorMethod fundMatrixEstimatorMethod, double[] qualityScores)
      Constructor.
      Parameters:
      fundMatrixEstimatorMethod - method for non-robust fundamental matrix estimator.
      qualityScores - quality scores corresponding to each provided pair of matched points.
      Throws:
      IllegalArgumentException - if provided quality scores length is smaller than required size (i.e. 7 matched pair of points).

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixEstimatorMethod fundMatrixEstimatorMethod, double[] qualityScores, FundamentalMatrixRobustEstimatorListener listener)
      Constructor.
      Parameters:
      fundMatrixEstimatorMethod - method for non-robust fundamental matrix estimator.
      qualityScores - quality scores corresponding to each provided pair of matched points.
      listener - listener to be notified of events such as when estimation starts, ends or its progress significantly changes.
      Throws:
      IllegalArgumentException - if provided quality scores length is smaller than required size (i.e. 7 matched pair of points).

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixEstimatorMethod fundMatrixEstimatorMethod, double[] qualityScores, List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints)
      Constructor.
      Parameters:
      fundMatrixEstimatorMethod - method for non-robust fundamental matrix estimator.
      qualityScores - quality scores corresponding to each provided pair of matched points.
      leftPoints - 2D points on left view.
      rightPoints - 2D points on right view.
      Throws:
      IllegalArgumentException - if provided list of points or quality scores do not have the same length or their length is less than 7 points.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixEstimatorMethod fundMatrixEstimatorMethod, double[] qualityScores, List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints, FundamentalMatrixRobustEstimatorListener listener)
      Constructor.
      Parameters:
      fundMatrixEstimatorMethod - method for non-robust fundamental matrix estimator.
      qualityScores - quality scores corresponding to each provided pair of matched points.
      leftPoints - 2D points on left view.
      rightPoints - 2D points on right view.
      listener - listener to be notified of events such as when estimation starts, ends or its progress significantly changes.
      Throws:
      IllegalArgumentException - if provided list of points or quality scores do not have the same length or their length is less than 7 points.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator()
      Constructor.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(FundamentalMatrixRobustEstimatorListener listener)
      Constructor.
      Parameters:
      listener - listener to be notified of events such as when estimation starts, ends or its progress significantly changes.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints)
      Constructor.
      Parameters:
      leftPoints - 2D points on left view.
      rightPoints - 2D points on right view.
      Throws:
      IllegalArgumentException - if provided list of points do not have the same length or their length is less than 8 points.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints, FundamentalMatrixRobustEstimatorListener listener)
      Constructor.
      Parameters:
      leftPoints - 2D points on left view.
      rightPoints - 2D points on right view.
      listener - listener to be notified of events such as when estimation starts, ends or its progress significantly changes.
      Throws:
      IllegalArgumentException - if provided list of points do not have the same length or their length is less than 8 points.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(double[] qualityScores)
      Constructor.
      Parameters:
      qualityScores - quality scores corresponding to each provided pair of matched points.
      Throws:
      IllegalArgumentException - if provided quality scores length is smaller than required size (i.e. 7 matched pair of points).

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(double[] qualityScores, FundamentalMatrixRobustEstimatorListener listener)
      Constructor.
      Parameters:
      qualityScores - quality scores corresponding to each provided pair of matched points.
      listener - listener to be notified of events such as when estimation starts, ends or its progress significantly changes.
      Throws:
      IllegalArgumentException - if provided quality scores length is smaller than required size (i.e. 7 matched pair of points).

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(double[] qualityScores, List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints)
      Constructor.
      Parameters:
      qualityScores - quality scores corresponding to each provided pair of matched points.
      leftPoints - 2D points on left view.
      rightPoints - 2D points on right view.
      Throws:
      IllegalArgumentException - if provided list of points or quality scores do not have the same length or their length is less than 7 points.

    • PROMedSFundamentalMatrixRobustEstimator

      public PROMedSFundamentalMatrixRobustEstimator(double[] qualityScores, List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints, FundamentalMatrixRobustEstimatorListener listener)
      Constructor.
      Parameters:
      qualityScores - quality scores corresponding to each provided pair of matched points.
      leftPoints - 2D points on left view.
      rightPoints - 2D points on right view.
      listener - listener to be notified of events such as when estimation starts, ends or its progress significantly changes.
      Throws:
      IllegalArgumentException - if provided list of points or quality scores do not have the same length or their length is less than 7 points.

  • Method Details

    • setPoints

      public void setPoints(List<com.irurueta.geometry.Point2D> leftPoints, List<com.irurueta.geometry.Point2D> rightPoints) throws com.irurueta.geometry.estimators.LockedException
      Sets matched 2D points on both left and right views.
      Overrides:
      setPoints in class FundamentalMatrixRobustEstimator
      Parameters:
      leftPoints - matched 2D points on left view.
      rightPoints - matched 2D points on right view.
      Throws:
      com.irurueta.geometry.estimators.LockedException - if this fundamental matrix estimator is locked.
      IllegalArgumentException - if provided matched points on left and right views do not have the same length or if their length is less than 8 points.

    • getStopThreshold

      public double getStopThreshold()
      Returns threshold to be used to keep the algorithm iterating in case that best estimated threshold using median of residuals is not small enough. Once a solution is found that generates a threshold below this value, the algorithm will stop. The stop threshold can be used to prevent the LMedS algorithm iterating too many times in cases where samples have a very similar accuracy. For instance, in cases where proportion of outliers is very small (close to 0%), and samples are very accurate (i.e. 1e-6), the algorithm would iterate for a long time trying to find the best solution when indeed there is no need to do that if a reasonable threshold has already been reached. Because of this behaviour the stop threshold can be set to a value much lower than the one typically used in RANSAC, and yet the algorithm could still produce even smaller thresholds in estimated results.
      Returns:
      stop threshold to stop the algorithm prematurely when a certain accuracy has been reached.

    • setStopThreshold

      public void setStopThreshold(double stopThreshold) throws com.irurueta.geometry.estimators.LockedException
      Sets threshold to be used to keep the algorithm iterating in case that best estimated threshold using median of residuals is not small enough. Once a solution is found that generates a threshold below this value, the algorithm will stop. The stop threshold can be used to prevent the LMedS algorithm iterating too many times in cases where samples have a very similar accuracy. For instance, in cases where proportion of outliers is very small (close to 0%), and samples are very accurate (i.e. 1e-6), the algorithm would iterate for a long time trying to find the best solution when indeed there is no need to do that if a reasonable threshold has already been reached. Because of this behaviour the stop threshold can be set to a value much lower than the one typically used in RANSAC, and yet the algorithm could still produce even smaller thresholds in estimated results.
      Parameters:
      stopThreshold - stop threshold to stop the algorithm prematurely when a certain accuracy has been reached.
      Throws:
      IllegalArgumentException - if provided value is zero or negative.
      com.irurueta.geometry.estimators.LockedException - if robust estimator is locked because an estimation is already in progress.

    • getQualityScores

      public double[] getQualityScores()
      Returns quality scores corresponding to each provided pair of points. The larger the score value the better the quality of the sampled matched pair of points.
      Overrides:
      getQualityScores in class FundamentalMatrixRobustEstimator
      Returns:
      quality scores corresponding to each pair of points.

    • setQualityScores

      public void setQualityScores(double[] qualityScores) throws com.irurueta.geometry.estimators.LockedException
      Sets quality scores corresponding to each provided pair of points. The larger the score value the better the quality of the sampled matched pair of points.
      Overrides:
      setQualityScores in class FundamentalMatrixRobustEstimator
      Parameters:
      qualityScores - quality scores corresponding to each pair of points.
      Throws:
      com.irurueta.geometry.estimators.LockedException - if robust estimator is locked because an estimation is already in progress.
      IllegalArgumentException - if provided quality scores length is smaller than MINIMUM_SIZE (i.e. 3 samples).

    • isReady

      public boolean isReady()
      Returns value indicating whether required data has been provided so that fundamental matrix estimation can start. This is true when input data (i.e. 7 pairs of matched 2D points and their quality scores) are provided. If true, estimator is ready to compute a fundamental matrix, otherwise more data needs to be provided.
      Overrides:
      isReady in class FundamentalMatrixRobustEstimator
      Returns:
      true if estimator is ready, false otherwise.

    • estimate

      public FundamentalMatrix estimate() throws com.irurueta.geometry.estimators.LockedException, com.irurueta.geometry.estimators.NotReadyException, com.irurueta.numerical.robust.RobustEstimatorException
      Estimates a radial distortion using a robust estimator and the best set of matched 2D points found using the robust estimator.
      Specified by:
      estimate in class FundamentalMatrixRobustEstimator
      Returns:
      a radial distortion.
      Throws:
      com.irurueta.geometry.estimators.LockedException - if robust estimator is locked because an estimation is already in progress.
      com.irurueta.geometry.estimators.NotReadyException - if provided input data is not enough to start the estimation.
      com.irurueta.numerical.robust.RobustEstimatorException - if estimation fails for any reason (i.e. numerical instability, no solution available, etc).

    • getMethod

      public com.irurueta.numerical.robust.RobustEstimatorMethod getMethod()
      Returns method being used for robust estimation.
      Specified by:
      getMethod in class FundamentalMatrixRobustEstimator
      Returns:
      method being used for robust estimation.

    • getRefinementStandardDeviation

      protected double getRefinementStandardDeviation()
      Gets standard deviation used for Levenberg-Marquardt fitting during refinement. Returned value gives an indication of how much variance each residual has. Typically, this value is related to the threshold used on each robust estimation, since residuals of found inliers are within the range of such threshold.
      Specified by:
      getRefinementStandardDeviation in class FundamentalMatrixRobustEstimator
      Returns:
      standard deviation used for refinement.

    • internalSetQualityScores

      private void internalSetQualityScores(double[] qualityScores)
      Sets quality scores corresponding to each provided pair of matched points. This method is used internally and does not check whether instance is locked or not.
      Parameters:
      qualityScores - quality scores to be set.
      Throws:
      IllegalArgumentException - if provided quality scores length is smaller than 8 points.