def _mask_dataset():
# Load model and compile function
with open('../cache/foreground_model.pkl', 'rb') as f:
model, threshold = cPickle.load(f)
x = T.matrix('x')
y = model.get_output(x)
output_func = theano.function(inputs=[x], outputs=(y >= threshold))
# Load data
image_data = DataLoader('../data/cuhk_small.mat', verbose=True)
# Pre-processing
print "Pre-processing ..."
images = image_data.get_all_images()
images = [_input_preproc(image) for image in images]
images = imageproc.images2mat(images).astype(theano.config.floatX)
# Compute masks
print "Computing masks ..."
masks = output_func(images)
# Save masks
print "Saving data ..."
mask_data = DataSaver()
cur_index = 0
for gid in xrange(image_data.get_n_groups()):
m, v = image_data.get_n_pedes_views(gid)
mask_data.add_group(m, v)
for pid in xrange(m):
n_images = image_data.get_n_images(gid, pid)
for vid, n in enumerate(n_images):
view_masks = [0] * n
for k in xrange(n):
mask = masks[cur_index, :]
mask = mask.reshape(160, 80, 1)
orig_image = image_data.get_image(gid, pid, vid, k)
orig_image = imageproc.imresize(orig_image, (160, 80, 3))
view_masks[k] = (mask * orig_image).astype(numpy.uint8)
cur_index += 1
mask_data.set_images(gid, pid, vid, view_masks)
mask_data.save('../data/cuhk_small_masked.mat')
def _mask_dataset():
# Load model and compile function
with open('../cache/foreground_model.pkl', 'rb') as f:
model, threshold = cPickle.load(f)
x = T.matrix('x')
y = model.get_output(x)
output_func = theano.function(inputs=[x], outputs=(y >= threshold))
# Load data
image_data = DataLoader('../data/cuhk_small.mat', verbose=True)
# Pre-processing
print "Pre-processing ..."
images = image_data.get_all_images()
images = [_input_preproc(image) for image in images]
images = imageproc.images2mat(images).astype(theano.config.floatX)
# Compute masks
print "Computing masks ..."
masks = output_func(images)
# Save masks
print "Saving data ..."
mask_data = DataSaver()
cur_index = 0
for gid in xrange(image_data.get_n_groups()):
m, v = image_data.get_n_pedes_views(gid)
mask_data.add_group(m, v)
for pid in xrange(m):
n_images = image_data.get_n_images(gid, pid)
for vid, n in enumerate(n_images):
view_masks = [0] * n
for k in xrange(n):
mask = masks[cur_index, :]
mask = mask.reshape(160, 80, 1)
orig_image = image_data.get_image(gid, pid, vid, k)
orig_image = imageproc.imresize(orig_image, (160, 80, 3))
view_masks[k] = (mask * orig_image).astype(numpy.uint8)
cur_index += 1
mask_data.set_images(gid, pid, vid, view_masks)
mask_data.save('../data/cuhk_small_masked.mat')
def show_pid(self, pid):
# Show images
self._gallery.itemSelectionChanged.disconnect(self.update_attr)
self._gallery.clear()
for v in xrange(self.mat_images.shape[1]):
if self.mat_images[pid, v].size == 0: break
for k in xrange(self.mat_images[pid, v].shape[1]):
img = self.mat_images[pid, v][0, k]
img = ndarray2qimage(imageproc.imresize(img, (256, 96)))
item = QtGui.QListWidgetItem(
QtGui.QIcon(QtGui.QPixmap.fromImage(img)),
'{0},{1}'.format(v, k))
self._gallery.addItem(item)
item.setSelected(True)
self._gallery.itemSelectionChanged.connect(self.update_attr)
# Show attributes
self.update_attr()
# Show status
self._status.setText('{0} / {1}'.format(pid+1, self.mat_images.shape[0]))
def imgprep(img):
img = imageproc.imresize(img, (80, 30))
img = imageproc.subtract_luminance(img)
img = numpy.rollaxis(img, 2)
return img
def imgprep(img):
img = imageproc.imresize(img, (80, 30))
img = rgb2lab(img)[:,:,1:]
img = numpy.rollaxis(img, 2)
return img
def imgprep(img):
img = imageproc.imresize(img, (80, 30))
img = rgb2lab(img)[:, :, 1:]
img = numpy.rollaxis(img, 2)
return img
def _preproc(image):
image = imageproc.imresize(image, (80, 40, 3))
image = numpy.rollaxis(image, 2)
# image = imageproc.subtract_luminance(image)
# image = imageproc.scale_per_channel(image, [0, 1])
return image / 255.0
def _input_preproc(image):
image = imageproc.imresize(image, (160, 80, 3))
image = imageproc.subtract_luminance(image)
image = imageproc.scale_per_channel(image, [0, 1])
image = numpy.rollaxis(image, 2)
return image
def _preproc_func(image):
image = imageproc.imresize(image, (80, 40), keep_ratio='height')
image = imageproc.subtract_luminance(image)
image = numpy.rollaxis(image, 2)
return numpy.tanh(image)
def imgprep(img):
img = imageproc.imresize(img, (80, 30))
img = imageproc.subtract_luminance(img)
img = numpy.rollaxis(img, 2)
return img
def _preproc_func(image):
image = imageproc.imresize(image, (80, 40), keep_ratio='height')
image = imageproc.subtract_luminance(image)
image = numpy.rollaxis(image, 2)
return numpy.tanh(image)
def _preproc(image):
image = imageproc.imresize(image, (80, 40, 3))
image = numpy.rollaxis(image, 2)
# image = imageproc.subtract_luminance(image)
# image = imageproc.scale_per_channel(image, [0, 1])
return image / 255.0
def _input_preproc(image):
image = imageproc.imresize(image, (160, 80, 3))
image = imageproc.subtract_luminance(image)
image = imageproc.scale_per_channel(image, [0, 1])
image = numpy.rollaxis(image, 2)
return image
