diff --git a/bob/bio/face/baseline/baseline.py b/bob/bio/face/baseline/baseline.py
index e8a76e53fddb1c4237de035f85f07dbc5707ec28..cd7aae465d3b6041fc31ce4b4393b042080f2daa 100644
--- a/bob/bio/face/baseline/baseline.py
+++ b/bob/bio/face/baseline/baseline.py
@@ -8,11 +8,6 @@ Defining some face recognition baselines
from bob.bio.base.baseline import Baseline
-eigenface = Baseline(name="eigenface",
- preprocessors={'default': 'face-crop-eyes', 'atnt': 'base'},
- extractor='linearize',
- algorithm='pca')
-
lda = Baseline(name="lda",
preprocessors={'default': 'face-crop-eyes', 'atnt': 'base'},
extractor='eigenface',
diff --git a/bob/bio/face/config/extractor/eigenface.py b/bob/bio/face/config/extractor/eigenface.py
deleted file mode 100644
index f0f38a77964c32092b4a575ff46bd592138126b4..0000000000000000000000000000000000000000
--- a/bob/bio/face/config/extractor/eigenface.py
+++ /dev/null
@@ -1,8 +0,0 @@
-#!/usr/bin/env python
-
-import bob.bio.face
-
-# compute eigenfaces using the training database
-extractor = bob.bio.face.extractor.Eigenface(
- subspace_dimension = .95
-)
diff --git a/bob/bio/face/extractor/Eigenface.py b/bob/bio/face/extractor/Eigenface.py
deleted file mode 100644
index c0dc456c21fb41ac936b9a491d4489267f2b4c10..0000000000000000000000000000000000000000
--- a/bob/bio/face/extractor/Eigenface.py
+++ /dev/null
@@ -1,112 +0,0 @@
-#!/usr/bin/env python
-# vim: set fileencoding=utf-8 :
-# Manuel Guenther <Manuel.Guenther@idiap.ch>
-
-import numpy
-
-import bob.learn.linear
-import bob.io.base
-
-from bob.bio.base.extractor import Extractor
-
-import logging
-logger = logging.getLogger("bob.bio.face")
-
-class Eigenface (Extractor):
- """Performs a principal component analysis (PCA) on the given data.
-
- This algorithm computes a PCA projection (:py:class:`bob.learn.linear.PCATrainer`) on the given training images, and projects the images into face space.
- In opposition to :py:class:`bob.bio.base.algorithm.PCA`, here the eigenfces are used as features, i.e., to apply advanced face recognition algorithms on top of them.
-
- **Parameters:**
-
- subspace_dimension : int or float
- If specified as ``int``, defines the number of eigenvectors used in the PCA projection matrix.
- If specified as ``float`` (between 0 and 1), the number of eigenvectors is calculated such that the given percentage of variance is kept.
-
- kwargs : ``key=value`` pairs
- A list of keyword arguments directly passed to the :py:class:`bob.bio.base.extractor.Extractor` base class constructor.
- """
-
- def __init__(self, subspace_dimension):
- # We have to register that this function will need a training step
- Extractor.__init__(self, requires_training = True, subspace_dimension = subspace_dimension)
- self.subspace_dimension = subspace_dimension
-
-
- def _check_data(self, data):
- """Checks that the given data are appropriate."""
- assert isinstance(data, numpy.ndarray)
- assert data.ndim == 2
- assert data.dtype == numpy.float64
-
-
- def train(self, training_images, extractor_file):
- """Generates the PCA covariance matrix and writes it into the given extractor_file.
-
- Beforehand, all images are turned into a 1D pixel vector.
-
- **Parameters:**
-
- training_images : [2D :py:class:`numpy.ndarray`]
- A list of 2D training images to train the PCA projection matrix with.
-
- extractor_file : str
- A writable file, into which the PCA projection matrix (as a :py:class:`bob.learn.linear.Machine`) will be written.
- """
- [self._check_data(image) for image in training_images]
-
- # Initializes an array for the data
- data = numpy.vstack([image.flatten() for image in training_images])
-
- logger.info(" -> Training LinearMachine using PCA (SVD)")
- t = bob.learn.linear.PCATrainer()
- self.machine, variances = t.train(data)
-
- # compute variance percentage, if desired
- if isinstance(self.subspace_dimension, float):
- cummulated = numpy.cumsum(variances) / numpy.sum(variances)
- for index in range(len(cummulated)):
- if cummulated[index] > self.subspace_dimension:
- self.subspace_dimension = index
- break
- self.subspace_dimension = index
- logger.info(" -> Keeping %d eigenvectors" % self.subspace_dimension)
-
- # Machine: get shape, then resize
- self.machine.resize(self.machine.shape[0], self.subspace_dimension)
- self.machine.save(bob.io.base.HDF5File(extractor_file, "w"))
-
-
- def load(self, extractor_file):
- """Reads the PCA projection matrix from file.
-
- **Parameters:**
-
- extractor_file : str
- An existing file, from which the PCA projection matrix are read.
- """
- # read PCA projector
- self.machine = bob.learn.linear.Machine(bob.io.base.HDF5File(extractor_file))
-
-
- def __call__(self, image):
- """__call__(image) -> feature
-
- Projects the given image using the stored covariance matrix.
-
- Beforehand, the image is turned into a 1D pixel vector.
-
- **Parameters:**
-
- image : 2D :py:class:`numpy.ndarray` (floats)
- The image to extract the eigenface feature from.
-
- **Returns:**
-
- feature : 1D :py:class:`numpy.ndarray` (floats)
- The extracted eigenface feature.
- """
- self._check_data(image)
- # Projects the data
- return self.machine(image.flatten())
diff --git a/bob/bio/face/test/test_extractors.py b/bob/bio/face/test/test_extractors.py
index 4effa7c6c83ab1cfbf38f42a3aaf9547d4db82e3..65b9dbd23696c0e555f135c83f98d2e52406ba4f 100644
--- a/bob/bio/face/test/test_extractors.py
+++ b/bob/bio/face/test/test_extractors.py
@@ -153,44 +153,6 @@ def test_lgbphs():
_compare(feature, reference, lgbphs.write_feature, lgbphs.read_feature)
-def test_eigenface():
- temp_file = bob.io.base.test_utils.temporary_filename()
- data = _data()
- eigen1 = bob.bio.base.load_resource('eigenface', 'extractor', preferred_package='bob.bio.face')
- assert isinstance(eigen1, bob.bio.face.extractor.Eigenface)
- assert isinstance(eigen1, bob.bio.base.extractor.Extractor)
- assert eigen1.requires_training
-
- # create extractor with a smaller number of kept eigenfaces
- train_data = utils.random_training_set(data.shape, 400, 0., 255.)
- eigen2 = bob.bio.face.extractor.Eigenface(subspace_dimension = 5)
- reference = pkg_resources.resource_filename('bob.bio.face.test', 'data/eigenface_extractor.hdf5')
- try:
- # train the projector
- eigen2.train(train_data, temp_file)
-
- assert os.path.exists(temp_file)
-
- if regenerate_refs: shutil.copy(temp_file, reference_file)
-
- # check projection matrix
- eigen1.load(reference)
- eigen2.load(temp_file)
-
- assert eigen1.machine.shape == eigen2.machine.shape
- for i in range(5):
- assert numpy.abs(eigen1.machine.weights[:,i] - eigen2.machine.weights[:,i] < 1e-5).all() or numpy.abs(eigen1.machine.weights[:,i] + eigen2.machine.weights[:,i] < 1e-5).all()
-
- finally:
- if os.path.exists(temp_file): os.remove(temp_file)
-
- # now, we can execute the extractor and check that the feature is still identical
- feature = eigen1(data)
- assert feature.ndim == 1
- reference = pkg_resources.resource_filename('bob.bio.face.test', 'data/eigenface_feature.hdf5')
- _compare(feature, reference, eigen1.write_feature, eigen1.read_feature)
-
-
"""
def test05_sift_key_points(self):
# check if VLSIFT is available
diff --git a/setup.py b/setup.py
index 942080a27c247e31f14daf3f717e8fa36b3fe0a1..7a32a7a59cb22bea145bc4aa67cf0d8fe12b2745 100644
--- a/setup.py
+++ b/setup.py
@@ -185,7 +185,6 @@ setup(
'dct-blocks = bob.bio.face.config.extractor.dct_blocks:extractor', # DCT blocks
'grid-graph = bob.bio.face.config.extractor.grid_graph:extractor', # Grid graph
'lgbphs = bob.bio.face.config.extractor.lgbphs:extractor', # LGBPHS
- 'eigenface = bob.bio.face.config.extractor.eigenface:extractor', # Eigenface
],
'bob.bio.algorithm': [
@@ -196,7 +195,6 @@ setup(
#baselines
'bob.bio.baseline':[
- 'eigenface = bob.bio.face.baseline.baseline:eigenface',
'lda = bob.bio.face.baseline.baseline:lda',
'plda = bob.bio.face.baseline.baseline:plda',
'gabor_graph = bob.bio.face.baseline.baseline:gabor_graph',