

In [ ]:

    
%matplotlib inline
%pylab inline

Load training data



In [ ]:

    
import menpo.io as mio
from menpo.landmark import labeller, lfpw_face
from menpofast.utils import convert_from_menpo

path = '/data/'
group = 'lfpw_face'

training_images = []
for i in mio.import_images(path + 'PhD/DataBases/faces/cofw/trainset/',
                           verbose=True, max_images=None):
    
    # convert the image from menpo Image to menpofast Image (channels at front)
    i = convert_from_menpo(i)
    
    labeller(i, 'PTS', eval(group))
    i.crop_to_landmarks_proportion_inplace(1, group=group)
    i = i.rescale_landmarks_to_diagonal_range(200, group=group)
    
    if i.n_channels == 3:
        i = i.as_greyscale(mode='average')
    training_images.append(i)



In [3]:

    
from menpo.visualize import visualize_images

visualize_images(training_images)

Constrained Local Models

Build



In [4]:

    
from menpofast.feature import no_op, fast_dsift, fast_daisy
from alabortcvpr2015.clm import CLMBuilder
from alabortcvpr2015.clm.classifier import MCF

clm = CLMBuilder(parts_shape=(21, 21),
                 features=fast_dsift,
                 diagonal=100,
                 classifier=MCF,
                 normalize_parts=False,
                 covariance=3,
                 scales=(1, .5)).build(training_images, 
                                       group=group, 
                                       verbose=True)









    



  - Level 0: Done
  - Level 1: Done



In [5]:

    
clm.parts_filters()[1][8].view()









    Out[5]:





<menpo.visualize.viewmatplotlib.MatplotlibImageSubplotsViewer2d at 0x7fa623be6d50>

Save



In [6]:

    
from alabortcvpr2015.utils import pickle_dump

pickle_dump(clm, path + 'PhD/Models/clm_cofw_fast_dsift')