def __init__(self, min_coords, max_coords=None, force_valid=False):
if isinstance(min_coords, Dlt):
self.dlt_tfm = min_coords
else:
if max_coords is None:
max_coords = min_coords
min_coords = np.zeros(dim)
self.dlt_tfm = Dlt(offset=(max_coords + min_coords) / 2,
scale=(max_coords - min_coords) / 2)
if force_valid:
min_coords = self.min_coords
max_coords = self.max_coords
self.dlt_tfm = Dlt(offset=(max_coords + min_coords) / 2,
scale=(max_coords - min_coords) / 2)
def mean(self):
'''Returns the mean vector.'''
if self._mean is None:
self._mean = np.zeros(
self.ndim) if abs(self.m0) < EPSILON else self.m1 / self.m0
return self._mean
def __init__(self, offset=np.zeros(2), scale=1, angle=0, on=False):
self.offset = offset
self.scale = scale
self.angle = angle
self.on = on
def __init__(self, offset=np.zeros(2), linear=Lin2d()):
self.offset = offset
self.linear = linear
def __init__(self, offset=np.zeros(3), scale=np.ones(3), check_shapes=True):
self.offset = offset
self.scale = scale
if check_shapes:
_ = self.dim # just to check the shapes
def __init__(self, weight=np.identity(3), bias=np.zeros(3), check_shapes=True):
self.weight = weight
self.bias = bias
if check_shapes:
_ = self.dim # just to check shapes
def __init__(self, offset=np.zeros(2), scale=1):
self.offset = offset
self.scale = scale
def __init__(self, offset=np.zeros(2), unitary=np.identity(2)):
self.offset = offset
self.unitary = unitary