def aam_build_benchmark(training_images, training_options=None, verbose=False):
r"""
Builds an AAM model.
Parameters
----------
training_images: list of :class:MaskedImage objects
A list of the training images.
training_options: dictionary, optional
A dictionary with the parameters that will be passed in the AAMBuilder
(:class:menpo.fitmultilevel.aam.AAMBuilder).
If None, the default options will be used.
This is an example of the dictionary with the default options:
training_options = {'group': 'PTS',
'feature_type': 'igo',
'transform': PiecewiseAffine,
'trilist': None,
'normalization_diagonal': None,
'n_levels': 3,
'downscale': 2,
'scaled_shape_models': True,
'pyramid_on_features': True,
'max_shape_components': None,
'max_appearance_components': None,
'boundary': 3,
'interpolator': 'scipy'
}
For an explanation of the options, please refer to the AAMBuilder
documentation.
Default: None
verbose: boolean, optional
If True, it prints information regarding the AAM training.
Default: False
Returns
-------
aam: :class:menpo.fitmultilevel.aam.AAM object
The trained AAM model.
"""
if verbose:
print('AAM Training:')
# parse options
if training_options is None:
training_options = {}
# group option
group = training_options.pop('group', None)
# trilist option
trilist = training_options.pop('trilist', None)
if trilist is not None:
labeller(training_images[0], 'PTS', trilist)
training_options['trilist'] = \
training_images[0].landmarks[trilist.__name__].lms.trilist
# build aam
aam = AAMBuilder(**training_options).build(training_images,
group=group,
verbose=verbose)
return aam
normalization_diagonal=150,
n_levels=2,
downscale=1.3,
pyramid_on_features=True,
noise_std=0.04,
rotation=False,
n_perturbations=2,
interpolator='scipy').train(training_images, group='PTS')
aam = AAMBuilder(
feature_type=sparse_hog,
transform=PiecewiseAffine,
trilist=training_images[0].landmarks['ibug_68_trimesh'].lms.trilist,
normalization_diagonal=150,
n_levels=3,
downscale=1.2,
scaled_shape_models=False,
pyramid_on_features=True,
max_shape_components=None,
max_appearance_components=3,
boundary=3,
interpolator='scipy').build(training_images, group='PTS')
sdm2 = SDAAMTrainer(aam,
regression_type=mlr,
regression_features=weights,
noise_std=0.04,
rotation=False,
n_perturbations=1,
update='compositional',
md_transform=OrthoMDTransform,
def test_n_levels_exception():
aam = AAMBuilder(n_levels=0).build(training_images, group='PTS')
def test_downscale_exception():
aam = AAMBuilder(downscale=1).build(training_images, group='PTS')
assert (aam.downscale == 1)
aam = AAMBuilder(downscale=0).build(training_images, group='PTS')
def test_verbose_mock(mock_stdout):
aam = AAMBuilder().build(training_images, group='PTS', verbose=True)
def test_feature_type_with_pyramid_on_features_exception():
aam = AAMBuilder(feature_type=['igo', sparse_hog]).build(training_images,
group='PTS')
def test_boundary_exception():
aam = AAMBuilder(boundary=-1).build(training_images, group='PTS')
def test_max_appearance_components_exception():
aam = AAMBuilder(max_appearance_components=[1, 2]).build(training_images,
group='PTS')
def test_max_shape_components_exception():
aam = AAMBuilder(max_shape_components=[1, 0.2, 'a']).build(training_images,
group='PTS')
def test_normalization_diagonal_exception():
aam = AAMBuilder(normalization_diagonal=100).build(training_images,
group='PTS')
assert (aam.appearance_models[0].n_features == 410)
aam = AAMBuilder(normalization_diagonal=10).build(training_images,
group='PTS')
for i in range(4):
im = mio.import_builtin_asset(filenames[i])
im.crop_to_landmarks_proportion_inplace(0.1)
labeller(im, 'PTS', ibug_68_trimesh)
if im.n_channels == 3:
im = im.as_greyscale(mode='luminosity')
training_images.append(im)
# build aams
aam1 = AAMBuilder(
feature_type=['igo', sparse_hog, None],
transform=PiecewiseAffine,
trilist=training_images[0].landmarks['ibug_68_trimesh'].lms.trilist,
normalization_diagonal=150,
n_levels=3,
downscale=2,
scaled_shape_models=False,
pyramid_on_features=False,
max_shape_components=[1, 2, 3],
max_appearance_components=[3, 3, 3],
boundary=3,
interpolator='scipy').build(training_images, group='PTS')
aam2 = AAMBuilder(feature_type=None,
transform=ThinPlateSplines,
trilist=None,
normalization_diagonal=None,
n_levels=2,
downscale=1.2,
scaled_shape_models=True,
pyramid_on_features=False,