diff --git a/Examples/Learning/CMakeLists.txt b/Examples/Learning/CMakeLists.txt
index 8d25fa6e2280808f545dda543b621b6165e55cb7..89d15d3743f07ae549c64f8f4901f4386229da47 100644
--- a/Examples/Learning/CMakeLists.txt
+++ b/Examples/Learning/CMakeLists.txt
@@ -33,23 +33,8 @@ add_executable(SOMExample SOMExample.cxx)
 target_link_libraries(SOMExample ${OTB_LIBRARIES})
 
 if(OTBLibSVM_LOADED)
-add_executable(SVMImageClassificationExample SVMImageClassificationExample.cxx)
-target_link_libraries(SVMImageClassificationExample ${OTB_LIBRARIES})
-
 add_executable(SVMImageEstimatorClassificationMultiExample SVMImageEstimatorClassificationMultiExample.cxx)
 target_link_libraries(SVMImageEstimatorClassificationMultiExample ${OTB_LIBRARIES})
-
-add_executable(SVMImageModelEstimatorExample SVMImageModelEstimatorExample.cxx)
-target_link_libraries(SVMImageModelEstimatorExample ${OTB_LIBRARIES})
-
-add_executable(SVMPointSetClassificationExample SVMPointSetClassificationExample.cxx)
-target_link_libraries(SVMPointSetClassificationExample ${OTB_LIBRARIES})
-
-add_executable(SVMPointSetExample SVMPointSetExample.cxx)
-target_link_libraries(SVMPointSetExample ${OTB_LIBRARIES})
-
-add_executable(SVMPointSetModelEstimatorExample SVMPointSetModelEstimatorExample.cxx)
-target_link_libraries(SVMPointSetModelEstimatorExample ${OTB_LIBRARIES})
 endif()
 
 if(OTBOpenCV_LOADED)
diff --git a/Examples/Learning/SVMImageClassificationExample.cxx b/Examples/Learning/SVMImageClassificationExample.cxx
deleted file mode 100644
index c73ad425c0f6491abb3414d383808cfac3ec15e5..0000000000000000000000000000000000000000
--- a/Examples/Learning/SVMImageClassificationExample.cxx
+++ /dev/null
@@ -1,337 +0,0 @@
-/*
- * Copyright (C) 2005-2017 Centre National d'Etudes Spatiales (CNES)
- *
- * This file is part of Orfeo Toolbox
- *
- *     https://www.orfeo-toolbox.org/
- *
- * 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.
- */
-
-
-
-#include <fstream>
-
-#include "otbImageFileReader.h"
-#include "otbImageFileWriter.h"
-#include "itkUnaryFunctorImageFilter.h"
-#include "itkRescaleIntensityImageFilter.h"
-#include "otbImage.h"
-
-//  Software Guide : BeginCommandLineArgs
-//    INPUTS: {ROI_QB_MUL_1.png}
-//    OUTPUTS: {ROI_QB_MUL_1_SVN_CLASS.png}
-//    ${OTB_DATA_ROOT}/Examples/svm_image_model.svm
-//  Software Guide : EndCommandLineArgs
-
-//  Software Guide : BeginLatex
-// This example illustrates the use of the
-// \doxygen{otb}{SVMClassifier} class for performing SVM
-// classification on images.
-// In this example, we will use an SVM model estimated in the example
-// of section \ref{sec:LearningWithImages}
-// to separate between water and non-water pixels by using the RGB
-// values only. The images used for this example are shown in
-// figure~\ref{fig:SVMROIS}.
-// The first thing to do is include the header file for the
-// class. Since the \doxygen{otb}{SVMClassifier} takes
-// \doxygen{itk}{ListSample}s as input, the class
-// \doxygen{itk}{PointSetToListAdaptor} is needed.
-//
-//
-//  Software Guide : EndLatex
-
-#include "itkImageToListSampleAdaptor.h"
-
-// Software Guide : BeginCodeSnippet
-#include "otbSVMClassifier.h"
-// Software Guide : EndCodeSnippet
-
-int main(int argc, char* argv[])
-{
-
-  if (argc != 4)
-    {
-    std::cout << "Usage : " << argv[0] << " inputImage outputImage modelFile "
-              << std::endl;
-    return EXIT_FAILURE;
-    }
-
-  const char * imageFilename  = argv[1];
-  const char * modelFilename  = argv[3];
-  const char * outputFilename = argv[2];
-
-// Software Guide : BeginLatex
-//
-// In the framework of supervised learning and classification, we will
-// always use feature vectors for the characterization of the
-// classes. On the other hand, the class labels are scalar
-// values. Here, we start by defining the type of the features as the
-// \code{PixelType}, which will be used to define the feature
-// \code{VectorType}. We also declare the type for the labels.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef double                 PixelType;
-  typedef int                    LabelPixelType;
-// Software Guide : EndCodeSnippet
-  const unsigned int Dimension = 2;
-
-// Software Guide : BeginLatex
-//
-// We can now proceed to define the image type used for storing the
-// features. We also define the reader.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef otb::Image<itk::FixedArray<PixelType, 3>,
-      Dimension>          InputImageType;
-
-  typedef otb::ImageFileReader<InputImageType> ReaderType;
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We can now read the image by calling the \code{Update} method of the reader.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  ReaderType::Pointer reader = ReaderType::New();
-
-  reader->SetFileName(imageFilename);
-
-  reader->Update();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// The image has now to be transformed to a sample which
-// is compatible with the classification framework. We will use a
-// \doxygen{itk}{Statistics::ImageToListSampleAdaptor} for this
-// task. This class is templated over the image type used for
-// storing the measures.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef itk::Statistics::ImageToListSampleAdaptor<InputImageType> SampleType;
-  SampleType::Pointer sample = SampleType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// After instantiation, we can set the image as an imput of our
-// sample adaptor.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  sample->SetImage(reader->GetOutput());
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// Now, we need to declare the SVM model which is to be used by the
-// classifier. The SVM model is templated over the type of value used
-// for the measures and the type of pixel used for the labels.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef otb::SVMModel<PixelType, LabelPixelType> ModelType;
-
-  ModelType::Pointer model = ModelType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// After instantiation, we can load a model saved to a file (see
-// section \ref{sec:LearningWithImages} for an example of model
-// estimation and storage to a file.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  model->LoadModel(modelFilename);
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We have now all the elements to create a classifier. The classifier
-// is templated over the sample type (the type of the data to be
-// classified) and the label type (the type of the output of the classifier).
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef otb::SVMClassifier<SampleType, LabelPixelType> ClassifierType;
-
-  ClassifierType::Pointer classifier = ClassifierType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We set the classifier parameters : number of classes, SVM model,
-// the sample data. And we trigger the classification process by
-// calling the \code{Update} method.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  int numberOfClasses = model->GetNumberOfClasses();
-  classifier->SetNumberOfClasses(numberOfClasses);
-  classifier->SetModel(model);
-  classifier->SetInput(sample.GetPointer());
-  classifier->Update();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// After the classification step, we usually want to get the
-// results. The classifier gives an output under the form of a sample
-// list. This list supports the classical STL iterators. Therefore, we
-// will create an output image and fill it up with the results of the
-// classification. The pixel type of the output image is the same as
-// the one used for the labels.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef ClassifierType::ClassLabelType         OutputPixelType;
-  typedef otb::Image<OutputPixelType, Dimension> OutputImageType;
-
-  OutputImageType::Pointer outputImage = OutputImageType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We allocate the memory for the output image using the information
-// from the input image.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef itk::Index<Dimension>       myIndexType;
-  typedef itk::Size<Dimension>        mySizeType;
-  typedef itk::ImageRegion<Dimension> myRegionType;
-
-  mySizeType size;
-  size[0] = reader->GetOutput()->GetRequestedRegion().GetSize()[0];
-  size[1] = reader->GetOutput()->GetRequestedRegion().GetSize()[1];
-
-  myIndexType start;
-  start[0] = 0;
-  start[1] = 0;
-
-  myRegionType region;
-  region.SetIndex(start);
-  region.SetSize(size);
-
-  outputImage->SetRegions(region);
-  outputImage->Allocate();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We can now declare the iterators on the list that we get at the
-// output of the classifier as well as the iterator to fill the output image.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  ClassifierType::OutputType* membershipSample =
-    classifier->GetOutput();
-  ClassifierType::OutputType::ConstIterator m_iter =
-    membershipSample->Begin();
-  ClassifierType::OutputType::ConstIterator m_last =
-    membershipSample->End();
-
-  typedef itk::ImageRegionIterator<OutputImageType> OutputIteratorType;
-  OutputIteratorType outIt(outputImage,
-                           outputImage->GetBufferedRegion());
-
-  outIt.GoToBegin();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We will iterate through the list, get the labels and assign pixel
-// values to the output image.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  while (m_iter != m_last && !outIt.IsAtEnd())
-    {
-    outIt.Set(m_iter.GetClassLabel());
-    ++m_iter;
-    ++outIt;
-    }
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// Only for visualization purposes, we choose to rescale the image of
-// classes before saving it to a file. We will use the
-// \doxygen{itk}{RescaleIntensityImageFilter} for this purpose.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef otb::Image<unsigned char, Dimension> FileImageType;
-
-  typedef itk::RescaleIntensityImageFilter<OutputImageType,
-      FileImageType> RescalerType;
-
-  RescalerType::Pointer rescaler = RescalerType::New();
-
-  rescaler->SetOutputMinimum(itk::NumericTraits<unsigned char>::min());
-  rescaler->SetOutputMaximum(itk::NumericTraits<unsigned char>::max());
-
-  rescaler->SetInput(outputImage);
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We can now create an image file writer and save the image.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef otb::ImageFileWriter<FileImageType> WriterType;
-
-  WriterType::Pointer writer = WriterType::New();
-
-  writer->SetFileName(outputFilename);
-  writer->SetInput(rescaler->GetOutput());
-
-  writer->Update();
-// Software Guide : EndCodeSnippet
-
-//  Software Guide : BeginLatex
-// Figure \ref{fig:SVMCLASS} shows the result of the SVM classification.
-// \begin{figure}
-// \center
-// \includegraphics[width=0.45\textwidth]{ROI_QB_MUL_1.eps}
-// \includegraphics[width=0.45\textwidth]{ROI_QB_MUL_1_SVN_CLASS.eps}
-// \itkcaption[SVM Image Classification]{Result of the SVM
-// classification . Left: RGB image. Right: image of classes.}
-// \label{fig:SVMCLASS}
-// \end{figure}
-//  Software Guide : EndLatex
-
-  return EXIT_SUCCESS;
-}
diff --git a/Examples/Learning/SVMImageModelEstimatorExample.cxx b/Examples/Learning/SVMImageModelEstimatorExample.cxx
deleted file mode 100644
index 9ee96db859d87531a99b575f86758e05c6e4e360..0000000000000000000000000000000000000000
--- a/Examples/Learning/SVMImageModelEstimatorExample.cxx
+++ /dev/null
@@ -1,161 +0,0 @@
-/*
- * Copyright (C) 2005-2017 Centre National d'Etudes Spatiales (CNES)
- *
- * This file is part of Orfeo Toolbox
- *
- *     https://www.orfeo-toolbox.org/
- *
- * 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.
- */
-
-
-
-//  Software Guide : BeginCommandLineArgs
-//    INPUTS: {ROI_QB_MUL_1.png}, {ROI_mask.png}
-//    OUTPUTS: {svm_image_model.svn}
-//  Software Guide : EndCommandLineArgs
-
-//  Software Guide : BeginLatex
-// This example illustrates the use of the
-// \doxygen{otb}{SVMImageModelEstimator} class. This class allows the
-// estimation of a SVM model (supervised learning) from a feature
-// image and an image of labels. In this example, we will train an SVM
-// to separate between water and non-water pixels by using the RGB
-// values only. The images used for this example are shown in
-// figure~\ref{fig:SVMROIS}.
-// \begin{figure}
-// \center
-// \includegraphics[width=0.45\textwidth]{ROI_QB_MUL_1.eps}
-// \includegraphics[width=0.45\textwidth]{ROI_mask.eps}
-// \itkcaption[SVM Image Model Estimation]{Images used for the
-// estimation of the SVM model. Left: RGB image. Right: image of labels.}
-// \label{fig:SVMROIS}
-// \end{figure}
-// The first thing to do is include the header file for the class.
-//
-//  Software Guide : EndLatex
-
-#include "itkMacro.h"
-#include "otbImage.h"
-#include "otbVectorImage.h"
-#include <iostream>
-
-//  Software Guide : BeginCodeSnippet
-#include "otbSVMImageModelEstimator.h"
-//  Software Guide : EndCodeSnippet
-
-#include "otbImageFileReader.h"
-
-int main(int itkNotUsed(argc), char* argv[])
-{
-
-  const char* inputImageFileName = argv[1];
-  const char* trainingImageFileName = argv[2];
-  const char* outputModelFileName = argv[3];
-
-//  Software Guide : BeginLatex
-//
-//  We define the types for the input and training images. Even if the
-//  input image will be an RGB image, we can read it as a 3 component
-//  vector image. This simplifies the interfacing with OTB's SVM
-//  framework.
-//
-// Software Guide : EndLatex
-//  Software Guide : BeginCodeSnippet
-  typedef unsigned char InputPixelType;
-  const unsigned int Dimension = 2;
-
-  typedef otb::VectorImage<InputPixelType,  Dimension> InputImageType;
-
-  typedef otb::Image<InputPixelType,  Dimension> TrainingImageType;
-//  Software Guide : EndCodeSnippet
-
-//  Software Guide : BeginLatex
-//
-//  The \doxygen{otb}{SVMImageModelEstimator} class is templated over
-//  the input (features) and the training (labels) images.
-//
-// Software Guide : EndLatex
-//  Software Guide : BeginCodeSnippet
-  typedef otb::SVMImageModelEstimator<InputImageType,
-      TrainingImageType>   EstimatorType;
-
-//  Software Guide : EndCodeSnippet
-
-//  Software Guide : BeginLatex
-//
-//  As usual, we define the readers for the images.
-//
-// Software Guide : EndLatex
-//  Software Guide : BeginCodeSnippet
-  typedef otb::ImageFileReader<InputImageType>    InputReaderType;
-  typedef otb::ImageFileReader<TrainingImageType> TrainingReaderType;
-
-  InputReaderType::Pointer    inputReader = InputReaderType::New();
-  TrainingReaderType::Pointer trainingReader = TrainingReaderType::New();
-
-//  Software Guide : EndCodeSnippet
-
-//  Software Guide : BeginLatex
-//
-//  We read the images. It is worth to note that, in order to ensure
-//  the pipeline coherence, the output of the objects which precede the
-//  model estimator in the pipeline, must be up to date, so we call
-//  the corresponding \code{Update} methods.
-//
-// Software Guide : EndLatex
-//  Software Guide : BeginCodeSnippet
-  inputReader->SetFileName(inputImageFileName);
-  trainingReader->SetFileName(trainingImageFileName);
-
-  inputReader->Update();
-  trainingReader->Update();
-
-//  Software Guide : EndCodeSnippet
-
-//  Software Guide : BeginLatex
-//
-//  We can now instantiate the model estimator and set its parameters.
-//
-// Software Guide : EndLatex
-//  Software Guide : BeginCodeSnippet
-  EstimatorType::Pointer svmEstimator = EstimatorType::New();
-
-  svmEstimator->SetInputImage(inputReader->GetOutput());
-  svmEstimator->SetTrainingImage(trainingReader->GetOutput());
-
-//  Software Guide : EndCodeSnippet
-
-//  Software Guide : BeginLatex
-//
-//  The model estimation procedure is triggered by calling the
-//  estimator's \code{Update} method.
-//
-// Software Guide : EndLatex
-//  Software Guide : BeginCodeSnippet
-  svmEstimator->Update();
-
-//  Software Guide : EndCodeSnippet
-
-//  Software Guide : BeginLatex
-//
-//  Finally, the estimated model can be saved to a file for later use.
-//
-//  Software Guide : EndLatex
-//  Software Guide : BeginCodeSnippet
-  svmEstimator->SaveModel(outputModelFileName);
-
-//  Software Guide : EndCodeSnippet
-
-  return EXIT_SUCCESS;
-}
diff --git a/Examples/Learning/SVMPointSetClassificationExample.cxx b/Examples/Learning/SVMPointSetClassificationExample.cxx
deleted file mode 100644
index e092f3ec4e9036e3a0daf5a561c896806019403d..0000000000000000000000000000000000000000
--- a/Examples/Learning/SVMPointSetClassificationExample.cxx
+++ /dev/null
@@ -1,322 +0,0 @@
-/*
- * Copyright (C) 2005-2017 Centre National d'Etudes Spatiales (CNES)
- *
- * This file is part of Orfeo Toolbox
- *
- *     https://www.orfeo-toolbox.org/
- *
- * 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.
- */
-
-
-
-//  Software Guide : BeginCommandLineArgs
-//    INPUTS: {svm_model.svn}
-//    OUTPUTS:
-//  Software Guide : EndCommandLineArgs
-
-#include "itkMacro.h"
-#include <iostream>
-#include <cstdlib>
-
-//  Software Guide : BeginLatex
-// This example illustrates the use of the
-// \doxygen{otb}{SVMClassifier} class for performing SVM
-// classification on pointsets.
-// The first thing to do is include the header file for the
-// class. Since the \doxygen{otb}{SVMClassifier} takes
-// \doxygen{itk}{ListSample}s as input, the class
-// \doxygen{itk}{PointSetToListSampleAdaptor} is needed.
-//
-// We start by including the needed header files.
-//
-//  Software Guide : EndLatex
-
-//  Software Guide : BeginCodeSnippet
-#include "itkPointSetToListSampleAdaptor.h"
-#include "otbSVMClassifier.h"
-//  Software Guide : EndCodeSnippet
-
-int main(int itkNotUsed(argc), char* argv[])
-{
-// Software Guide : BeginLatex
-//
-// In the framework of supervised learning and classification, we will
-// always use feature vectors for the characterization of the
-// classes. On the other hand, the class labels are scalar
-// values. Here, we start by defining the type of the features as the
-// \code{PixelType}, which will be used to define the feature
-// \code{VectorType}. We also declare the type for the labels.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef float InputPixelType;
-
-  typedef std::vector<InputPixelType> InputVectorType;
-  typedef int                         LabelPixelType;
-// Software Guide : EndCodeSnippet
-  const unsigned int Dimension = 2;
-
-// Software Guide : BeginLatex
-//
-// We can now proceed to define the point sets used for storing the
-// features and the labels.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef itk::PointSet<InputVectorType,  Dimension> MeasurePointSetType;
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We will need to get access to the data stored in the point sets, so
-// we define the appropriate for the points and the points containers
-// used by the point sets (see the section \ref{sec:PointSetSection}
-// for more information on how to use point sets).
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef MeasurePointSetType::PointType MeasurePointType;
-  typedef MeasurePointSetType::PointsContainer MeasurePointsContainer;
-
-  MeasurePointSetType::Pointer    tPSet = MeasurePointSetType::New();
-  MeasurePointsContainer::Pointer tCont = MeasurePointsContainer::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We need now to build the test set for the SVM. In this
-// simple example, we will build a SVM who classes points depending on
-// which side of the line $x=y$ they are located. We start by
-// generating 500 random points.
-//
-// Software Guide : EndLatex
-
-  srand(0);
-
-  unsigned int pointId;
-// Software Guide : BeginCodeSnippet
-  int lowest = 0;
-  int range = 1000;
-
-  for (pointId = 0; pointId < 100; pointId++)
-    {
-
-    MeasurePointType tP;
-
-    int x_coord = lowest + static_cast<int>(range * (rand() / (RAND_MAX + 1.0)));
-    int y_coord = lowest + static_cast<int>(range * (rand() / (RAND_MAX + 1.0)));
-
-    std::cout << "coords : " << x_coord << " " << y_coord << std::endl;
-    tP[0] = x_coord;
-    tP[1] = y_coord;
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We push the features in the vector after a normalization which is
-// useful for SVM convergence.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-    InputVectorType measure;
-    measure.push_back(static_cast<InputPixelType>((x_coord * 1.0 -
-                                                   lowest) / range));
-    measure.push_back(static_cast<InputPixelType>((y_coord * 1.0 -
-                                                   lowest) / range));
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// And we insert the points in the points container.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-    tCont->InsertElement(pointId, tP);
-    tPSet->SetPointData(pointId, measure);
-
-    }
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// After the loop, we set the points container to the point set.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  tPSet->SetPoints(tCont);
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// Once the pointset is ready, we must transform it to a sample which
-// is compatible with the classification framework. We will use a
-// \doxygen{itk}{Statistics::PointSetToListSampleAdaptor} for this
-// task. This class is templated over the point set type used for
-// storing the measures.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef itk::Statistics::PointSetToListSampleAdaptor<MeasurePointSetType>
-  SampleType;
-  SampleType::Pointer sample = SampleType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// After instantiation, we can set the point set as an imput of our
-// sample adaptor.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  sample->SetPointSet(tPSet);
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// Now, we need to declare the SVM model which is to be used by the
-// classifier. The SVM model is templated over the type of value used
-// for the measures and the type of pixel used for the labels.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef otb::SVMModel<SampleType::MeasurementVectorType::ValueType,
-      LabelPixelType> ModelType;
-
-  ModelType::Pointer model = ModelType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// After instantiation, we can load a model saved to a file (see
-// section \ref{sec:LearningWithPointSets} for an example of model
-// estimation and storage to a file).
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  model->LoadModel(argv[1]);
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We have now all the elements to create a classifier. The classifier
-// is templated over the sample type (the type of the data to be
-// classified) and the label type (the type of the output of the classifier).
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef otb::SVMClassifier<SampleType, LabelPixelType> ClassifierType;
-
-  ClassifierType::Pointer classifier = ClassifierType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We set the classifier parameters : number of classes, SVM model,
-// the sample data. And we trigger the classification process by
-// calling the \code{Update} method.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  int numberOfClasses = model->GetNumberOfClasses();
-  classifier->SetNumberOfClasses(numberOfClasses);
-  classifier->SetModel(model);
-  classifier->SetInput(sample.GetPointer());
-  classifier->Update();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// After the classification step, we usually want to get the
-// results. The classifier gives an output under the form of a sample
-// list. This list supports the classical STL iterators.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  ClassifierType::OutputType* membershipSample =
-    classifier->GetOutput();
-
-  ClassifierType::OutputType::ConstIterator m_iter =
-    membershipSample->Begin();
-  ClassifierType::OutputType::ConstIterator m_last =
-    membershipSample->End();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We will iterate through the list, get the labels and compute the
-// classification error.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  double error = 0.0;
-  pointId = 0;
-  while (m_iter != m_last)
-    {
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We get the label for each point.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-    ClassifierType::ClassLabelType label = m_iter.GetClassLabel();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// And we compare it to the corresponding one of the test set.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-    InputVectorType measure;
-
-    tPSet->GetPointData(pointId, &measure);
-
-    ClassifierType::ClassLabelType expectedLabel;
-    if (measure[0] < measure[1]) expectedLabel = -1;
-    else expectedLabel = 1;
-
-    double dist = fabs(measure[0] - measure[1]);
-
-    if (label != expectedLabel) error++;
-
-    std::cout << int(label) << "/" << int(expectedLabel) << " --- " << dist <<
-    std::endl;
-
-    ++pointId;
-    ++m_iter;
-    }
-
-  std::cout << "Error = " << error / pointId << " % " << std::endl;
-// Software Guide : EndCodeSnippet
-
-  return EXIT_SUCCESS;
-}
diff --git a/Examples/Learning/SVMPointSetExample.cxx b/Examples/Learning/SVMPointSetExample.cxx
deleted file mode 100644
index 6c60460cc3211434eb089be045569299c6c48b88..0000000000000000000000000000000000000000
--- a/Examples/Learning/SVMPointSetExample.cxx
+++ /dev/null
@@ -1,219 +0,0 @@
-/*
- * Copyright (C) 2005-2017 Centre National d'Etudes Spatiales (CNES)
- *
- * This file is part of Orfeo Toolbox
- *
- *     https://www.orfeo-toolbox.org/
- *
- * 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.
- */
-
-
-
-#include "itkMacro.h"
-#include <iostream>
-#include <cstdlib>
-
-#include "otbSVMPointSetModelEstimator.h"
-#include "itkPointSetToListSampleAdaptor.h"
-#include "otbSVMClassifier.h"
-
-int main(int itkNotUsed(argc), char* itkNotUsed(argv)[])
-{
-
-  typedef float InputPixelType;
-
-  typedef std::vector<InputPixelType> InputVectorType;
-  typedef int                         LabelPixelType;
-  const unsigned int Dimension = 2;
-
-  typedef itk::PointSet<InputVectorType,  Dimension> MeasurePointSetType;
-
-  typedef itk::PointSet<LabelPixelType,  Dimension> LabelPointSetType;
-
-  MeasurePointSetType::Pointer mPSet = MeasurePointSetType::New();
-  LabelPointSetType::Pointer   lPSet = LabelPointSetType::New();
-
-  typedef MeasurePointSetType::PointType MeasurePointType;
-  typedef LabelPointSetType::PointType   LabelPointType;
-
-  typedef MeasurePointSetType::PointsContainer MeasurePointsContainer;
-  typedef LabelPointSetType::PointsContainer   LabelPointsContainer;
-
-  MeasurePointsContainer::Pointer mCont = MeasurePointsContainer::New();
-  LabelPointsContainer::Pointer   lCont = LabelPointsContainer::New();
-
-  /* We learn the y>x | y<x boundary*/
-//  srand((unsigned)time(0));
-  srand(0);
-  int lowest = 0;
-  int range = 1000;
-
-  unsigned int pointId;
-
-  for (pointId = 0; pointId < 500; pointId++)
-    {
-
-    MeasurePointType mP;
-    LabelPointType   lP;
-
-    int x_coord = lowest + static_cast<int>(range * (rand() / (RAND_MAX + 1.0)));
-    int y_coord = lowest + static_cast<int>(range * (rand() / (RAND_MAX + 1.0)));
-
-    std::cout << "coords : " << x_coord << " " << y_coord << std::endl;
-    mP[0] = x_coord;
-    mP[1] = y_coord;
-
-    lP[0] = x_coord;
-    lP[1] = y_coord;
-
-    InputVectorType measure;
-    measure.push_back(static_cast<InputPixelType>((x_coord * 1.0 -
-                                                   lowest) / range));
-    measure.push_back(static_cast<InputPixelType>((y_coord * 1.0 -
-                                                   lowest) / range));
-
-    LabelPixelType label;
-
-    if (x_coord < y_coord) label = 0;
-    else label = 1;
-
-    std::cout << "Label : " << label << std::endl;
-    std::cout << "Measures : " << measure[0] << " " << measure[1] << std::endl;
-
-    mCont->InsertElement(pointId, mP);
-    mPSet->SetPointData(pointId, measure);
-
-    lCont->InsertElement(pointId, lP);
-    lPSet->SetPointData(pointId, label);
-
-    }
-
-  mPSet->SetPoints(mCont);
-  lPSet->SetPoints(lCont);
-
-  typedef otb::SVMPointSetModelEstimator<MeasurePointSetType,
-      LabelPointSetType>   EstimatorType;
-
-  EstimatorType::Pointer estimator = EstimatorType::New();
-
-  estimator->SetInputPointSet(mPSet);
-  estimator->SetTrainingPointSet(lPSet);
-
-  estimator->Update();
-
-  std::cout << "Saving model" << std::endl;
-  estimator->SaveModel("model.svm");
-
-  // Build the test set
-
-  MeasurePointSetType::Pointer    tPSet = MeasurePointSetType::New();
-  MeasurePointsContainer::Pointer tCont = MeasurePointsContainer::New();
-
-  for (pointId = 0; pointId < 100; pointId++)
-    {
-
-    MeasurePointType tP;
-
-    int x_coord = lowest + static_cast<int>(range * (rand() / (RAND_MAX + 1.0)));
-    int y_coord = lowest + static_cast<int>(range * (rand() / (RAND_MAX + 1.0)));
-
-    std::cout << "coords : " << x_coord << " " << y_coord << std::endl;
-    tP[0] = x_coord;
-    tP[1] = y_coord;
-
-    InputVectorType measure;
-    measure.push_back(static_cast<InputPixelType>((x_coord * 1.0 -
-                                                   lowest) / range));
-    measure.push_back(static_cast<InputPixelType>((y_coord * 1.0 -
-                                                   lowest) / range));
-
-    std::cout << "Measures : " << measure[0] << " " << measure[1] << std::endl;
-
-    tCont->InsertElement(pointId, tP);
-    tPSet->SetPointData(pointId, measure);
-
-    }
-
-  tPSet->SetPoints(tCont);
-
-  // Classify
-
-  typedef itk::Statistics::PointSetToListSampleAdaptor<MeasurePointSetType>
-  SampleType;
-  SampleType::Pointer sample = SampleType::New();
-  sample->SetPointSet(tPSet);
-
-  std::cout << "Sample set to Adaptor" << std::endl;
-
-  /** preparing classifier and decision rule object */
-  typedef otb::SVMModel<SampleType::MeasurementVectorType::ValueType,
-      LabelPixelType> ModelType;
-
-  ModelType::Pointer model = estimator->GetModel();
-
-  int numberOfClasses = model->GetNumberOfClasses();
-
-  std::cout << "Classification for " << numberOfClasses << " classes " <<
-  std::endl;
-
-  typedef otb::SVMClassifier<SampleType, LabelPixelType> ClassifierType;
-
-  ClassifierType::Pointer classifier = ClassifierType::New();
-
-  classifier->SetNumberOfClasses(numberOfClasses);
-  classifier->SetModel(model);
-  classifier->SetInput(sample.GetPointer());
-  classifier->Update();
-
-  /* Build the class map */
-  std::cout << "Output image creation" << std::endl;
-
-  std::cout << "classifier get output" << std::endl;
-  ClassifierType::OutputType* membershipSample =
-    classifier->GetOutput();
-  std::cout << "Sample iterators" << std::endl;
-  ClassifierType::OutputType::ConstIterator m_iter =
-    membershipSample->Begin();
-  ClassifierType::OutputType::ConstIterator m_last =
-    membershipSample->End();
-
-  double error = 0.0;
-  pointId = 0;
-  while (m_iter != m_last)
-    {
-    ClassifierType::ClassLabelType label = m_iter.GetClassLabel();
-
-    InputVectorType measure;
-
-    tPSet->GetPointData(pointId, &measure);
-
-    ClassifierType::ClassLabelType expectedLabel;
-    if (measure[0] < measure[1]) expectedLabel = 0;
-    else expectedLabel = 1;
-
-    double dist = fabs(measure[0] - measure[1]);
-
-    if (label != expectedLabel) error++;
-
-    std::cout << int(label) << "/" << int(expectedLabel) << " --- " << dist <<
-    std::endl;
-
-    ++pointId;
-    ++m_iter;
-    }
-
-  std::cout << "Error = " << error / pointId << std::endl;
-
-  return EXIT_SUCCESS;
-}
diff --git a/Examples/Learning/SVMPointSetModelEstimatorExample.cxx b/Examples/Learning/SVMPointSetModelEstimatorExample.cxx
deleted file mode 100644
index a351194730eb35b0c6103dde6d00df644cb72a63..0000000000000000000000000000000000000000
--- a/Examples/Learning/SVMPointSetModelEstimatorExample.cxx
+++ /dev/null
@@ -1,264 +0,0 @@
-/*
- * Copyright (C) 2005-2017 Centre National d'Etudes Spatiales (CNES)
- *
- * This file is part of Orfeo Toolbox
- *
- *     https://www.orfeo-toolbox.org/
- *
- * 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.
- */
-
-
-
-#include "itkMacro.h"
-#include "itkPointSet.h"
-#include <iostream>
-#include <cstdlib>
-
-// Software Guide : BeginLatex
-//
-// This example illustrates the use of the
-// \doxygen{otb}{SVMPointSetModelEstimator} in order to perform the
-// SVM learning from an \doxygen{itk}{PointSet} data structure.
-//
-// The first step required to use this filter is to include its header file.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-#include "otbSVMPointSetModelEstimator.h"
-// Software Guide : EndCodeSnippet
-
-int main(int itkNotUsed(argc), char* itkNotUsed(argv)[])
-{
-
-// Software Guide : BeginLatex
-//
-// In the framework of supervised learning and classification, we will
-// always use feature vectors for the characterization of the
-// classes. On the other hand, the class labels are scalar
-// values. Here, we start by defining the type of the features as the
-// \code{PixelType}, which will be used to define the feature
-// \code{VectorType}. We also declare the type for the labels.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef float                  PixelType;
-  typedef std::vector<PixelType> VectorType;
-  typedef int                    LabelPixelType;
-// Software Guide : EndCodeSnippet
-  const unsigned int Dimension = 2;
-
-// Software Guide : BeginLatex
-//
-// We can now proceed to define the point sets used for storing the
-// features and the labels.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef itk::PointSet<VectorType,  Dimension> FeaturePointSetType;
-
-  typedef itk::PointSet<LabelPixelType,  Dimension> LabelPointSetType;
-
-  FeaturePointSetType::Pointer fPSet = FeaturePointSetType::New();
-  LabelPointSetType::Pointer   lPSet = LabelPointSetType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We will need to get access to the data stored in the point sets, so
-// we define the appropriate for the points and the points containers
-// used by the point sets (see the section \ref{sec:PointSetSection}
-// for more information oin haw to use point sets).
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef FeaturePointSetType::PointType FeaturePointType;
-  typedef LabelPointSetType::PointType   LabelPointType;
-
-  typedef FeaturePointSetType::PointsContainer FeaturePointsContainer;
-  typedef LabelPointSetType::PointsContainer   LabelPointsContainer;
-
-  FeaturePointsContainer::Pointer fCont = FeaturePointsContainer::New();
-  LabelPointsContainer::Pointer   lCont = LabelPointsContainer::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We need now to build the training set for the SVM learning. In this
-// simple example, we will build a SVM who classes points depending on
-// which side of the line $x=y$ they are located. We start by
-// generating 500 random points.
-//
-// Software Guide : EndLatex
-
-  /* We learn the y>x | y<x boundary*/
-  srand(0);
-
-// Software Guide : BeginCodeSnippet
-  int lowest = 0;
-  int range = 1000;
-
-  for (unsigned int pointId = 0; pointId < 500; pointId++)
-    {
-
-    FeaturePointType fP;
-    LabelPointType   lP;
-
-    int x_coord = lowest + static_cast<int>(range * (rand() / (RAND_MAX + 1.0)));
-    int y_coord = lowest + static_cast<int>(range * (rand() / (RAND_MAX + 1.0)));
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We set the coordinates of the points. They are the same for the
-// feature vector and for the label.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-    fP[0] = x_coord;
-    fP[1] = y_coord;
-
-    lP[0] = x_coord;
-    lP[1] = y_coord;
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We push the features in the vector after a normalization which is
-// useful for SVM convergence.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-    VectorType feature;
-    feature.push_back(static_cast<PixelType>((x_coord * 1.0 - lowest) / range));
-    feature.push_back(static_cast<PixelType>((y_coord * 1.0 - lowest) / range));
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// We decide on the label for each point.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-    LabelPixelType label;
-
-    if (x_coord < y_coord) label = -1;
-    else label = 1;
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// And we insert the points in the points containers.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-    fCont->InsertElement(pointId, fP);
-    fPSet->SetPointData(pointId, feature);
-
-    lCont->InsertElement(pointId, lP);
-    lPSet->SetPointData(pointId, label);
-
-    }
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// After the loop, we set the points containers to the point sets.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  fPSet->SetPoints(fCont);
-  lPSet->SetPoints(lCont);
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// Up to now, we have only prepared the data for the SVM learning. We
-// can now create the SVM model estimator. This class is templated
-// over the feature and the label point set types.
-// \index{otb::SVMPointSetModelEstimator}
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  typedef otb::SVMPointSetModelEstimator<FeaturePointSetType,
-      LabelPointSetType>   EstimatorType;
-
-  EstimatorType::Pointer estimator = EstimatorType::New();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// The next step consists in setting the point sets for the estimator
-// and the number of classes for the model. The feture point set is
-// set using the \code{SetInputPointSet} and the label point set is
-// set with the \code{SetTrainingPointSet} method.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  estimator->SetInputPointSet(fPSet);
-  estimator->SetTrainingPointSet(lPSet);
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// The model estimation is triggered by calling the \code{Update}
-// method.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  estimator->Update();
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// Finally, we can save the result of the learning to a file.
-//
-// Software Guide : EndLatex
-
-// Software Guide : BeginCodeSnippet
-  estimator->SaveModel("svm_model.svm");
-// Software Guide : EndCodeSnippet
-
-// Software Guide : BeginLatex
-//
-// The \doxygen{otb}{otbSVMModel} class provides several accessors in
-// order to get some information about the result of the learning
-// step. For instance, one can get the number of support vectors kept
-// to define the separation surface by using the
-// \code{GetNumberOfSupportVectors()}. This can be very useful to
-// detect some kind of overlearning (the number of support vectors is
-// close to the number of examples). One can also get the SVs
-// themselves by calling the \code {GetSupportVectors()}. The $\alpha$
-// values for the support vectors can be accessed by using the
-// \code{GetAlpha()} method. Finally the \code{Evaluate()} method will
-// return the result of the classification of a sample and the
-// \code{EvaluateHyperplaneDistance()} will return the distance of
-// the sample to the separating surface (or surfaces in the case of
-// multi-class problems).
-//
-// Software Guide : EndLatex
-
-  return EXIT_SUCCESS;
-}
diff --git a/Examples/Learning/test/CMakeLists.txt b/Examples/Learning/test/CMakeLists.txt
index b5d2b6f1f83f91122ae15f7026ca6023e5e7b2fb..92581e8f1318ed2aa8b0575aae8dc2ee4be2d026 100644
--- a/Examples/Learning/test/CMakeLists.txt
+++ b/Examples/Learning/test/CMakeLists.txt
@@ -21,19 +21,7 @@
 set(BASELINE ${OTB_DATA_ROOT}/Baseline/Examples/Learning)
 set(INPUTDATA ${OTB_DATA_ROOT}/Examples)
 
-# ------- SVMImageClassificationExampleTest----------
-
 if(OTBLibSVM_LOADED)
-otb_add_test(NAME leTeSVMImageClassificationExampleTest COMMAND ${OTB_TEST_DRIVER}
-  --compare-n-images ${NOTOL} 1
-    ${BASELINE}/ROI_QB_MUL_1_SVN_CLASS.png
-    ${TEMP}/ROI_QB_MUL_1_SVN_CLASS.png
-  Execute $<TARGET_FILE:SVMImageClassificationExample>
-    ${INPUTDATA}/ROI_QB_MUL_1.png
-    ${TEMP}/ROI_QB_MUL_1_SVN_CLASS.png
-    ${OTB_DATA_ROOT}/Examples/svm_image_model.svm
-)
-
 # ------- SVMImageEstimatorClassificationMultiExampleTest----------
 
 otb_add_test(NAME leTeSVMImageEstimatorClassificationMultiExampleTest COMMAND ${OTB_TEST_DRIVER}
@@ -47,19 +35,6 @@ otb_add_test(NAME leTeSVMImageEstimatorClassificationMultiExampleTest COMMAND ${
     ${TEMP}/ROI_QB_MUL_1_SVN_CLASS_MULTI_Rescaled.png
 )
 
-# ------- SVMImageModelEstimatorExampleTest----------
-
-otb_add_test(NAME leTeSVMImageModelEstimatorExampleTest COMMAND ${OTB_TEST_DRIVER}
-  --compare-ascii ${EPSILON_3}
-    ${BASELINE}/svm_image_model.svn
-    ${TEMP}/svm_image_model.svn
-  --ignore-lines-with 2 probA probB
-  Execute $<TARGET_FILE:SVMImageModelEstimatorExample>
-    ${INPUTDATA}/ROI_QB_MUL_1.png
-    ${INPUTDATA}/ROI_mask.png
-    ${TEMP}/svm_image_model.svn
-)
-
 endif()
 # ------- GenerateTrainingImageExampleTest----------