from menpofit import checks
from menpofit.fitter import MultiScaleParametricFitter
from .algorithm import InverseCompositional
[docs]class LucasKanadeATMFitter(MultiScaleParametricFitter):
r"""
Class for defining an ATM fitter using the Lucas-Kanade optimization.
Parameters
----------
atm : :map:`ATM` or `subclass`
The trained ATM model.
lk_algorithm_cls : `class`, optional
The Lukas-Kanade optimisation algorithm that will get applied. The
possible algorithms are:
=========================== ============== =============
Class Warp Direction Warp Update
=========================== ============== =============
:map:`ForwardCompositional` Forward Compositional
:map:`InverseCompositional` Inverse
=========================== ============== =============
n_shape : `int` or `float` or `list` of those or ``None``, optional
The number of shape components that will be used. If `int`, then it
defines the exact number of active components. If `float`, then it
defines the percentage of variance to keep. If `int` or `float`, then
the provided value will be applied for all scales. If `list`, then it
defines a value per scale. If ``None``, then all the available
components will be used. Note that this simply sets the active
components without trimming the unused ones. Also, the available
components may have already been trimmed to `max_shape_components`
during training.
sampling : `list` of `int` or `ndarray` or ``None``
It defines a sampling mask per scale. If `int`, then it defines the
sub-sampling step of the sampling mask. If `ndarray`, then it
explicitly defines the sampling mask. If ``None``, then no
sub-sampling is applied.
"""
def __init__(self, atm, lk_algorithm_cls=InverseCompositional,
n_shape=None, sampling=None):
# Store model
self._model = atm
# Check parameters
checks.set_models_components(atm.shape_models, n_shape)
self._sampling = checks.check_sampling(sampling, atm.n_scales)
# Get list of algorithm objects per scale
interfaces = atm.build_fitter_interfaces(self._sampling)
algorithms = [lk_algorithm_cls(interface) for interface in interfaces]
# Call superclass
super(LucasKanadeATMFitter, self).__init__(
scales=atm.scales, reference_shape=atm.reference_shape,
holistic_features=atm.holistic_features, algorithms=algorithms)
@property
def atm(self):
r"""
The trained ATM model.
:type: :map:`ATM` or `subclass`
"""
return self._model
[docs] def warped_images(self, image, shapes):
r"""
Given an input test image and a list of shapes, it warps the image
into the shapes. This is useful for generating the warped images of a
fitting procedure stored within a
:map:`MultiScaleParametricIterativeResult`.
Parameters
----------
image : `menpo.image.Image` or `subclass`
The input image to be warped.
shapes : `list` of `menpo.shape.PointCloud`
The list of shapes in which the image will be warped. The shapes
are obtained during the iterations of a fitting procedure.
Returns
-------
warped_images : `list` of `menpo.image.MaskedImage` or `ndarray`
The warped images.
"""
return self.algorithms[-1].interface.warped_images(image=image,
shapes=shapes)
def __str__(self):
# Compute scale info strings
scales_info = []
lvl_str_tmplt = r""" - Scale {}
- {} active shape components
- {} similarity transform components"""
for k, s in enumerate(self.scales):
scales_info.append(lvl_str_tmplt.format(
s,
self.atm.shape_models[k].n_active_components,
self.atm.shape_models[k].n_global_parameters))
scales_info = '\n'.join(scales_info)
cls_str = r"""{class_title}
- Scales: {scales}
{scales_info}
""".format(class_title=self.algorithms[0].__str__(),
scales=self.scales,
scales_info=scales_info)
return self.atm.__str__() + cls_str