def test_linear_coefficients_jacobian_random(self, size):
"""Tests the Jacobian of the linear_coefficients function."""
tensor_shape = np.random.randint(size, 6, size=3)
matte_init = np.random.uniform(0.0,
1.0,
size=tensor_shape.tolist() + [1])
tensor_shape[1:3] -= (size - 1)
num_coeffs = np.random.randint(2, 4)
pseudo_inverse_init = np.random.uniform(0.0,
1.0,
size=tensor_shape.tolist() +
[num_coeffs, size**2])
matte = tf.convert_to_tensor(value=matte_init)
pseudo_inverse = tf.convert_to_tensor(value=pseudo_inverse_init)
a, b = matting.linear_coefficients(matte, pseudo_inverse)
with self.subTest(name="matte_a"):
self.assert_jacobian_is_correct(matte, matte_init, a)
with self.subTest(name="matte_b"):
self.assert_jacobian_is_correct(matte, matte_init, b)
with self.subTest(name="pseudo_inverse_a"):
self.assert_jacobian_is_correct(pseudo_inverse,
pseudo_inverse_init, a)
with self.subTest(name="pseudo_inverse_b"):
self.assert_jacobian_is_correct(pseudo_inverse,
pseudo_inverse_init, b)
def test_linear_coefficients_reconstruction_opposite_images(self, size):
"""Tests that the matte can be reconstructed by using the coefficients ."""
tensor_shape = np.random.randint(size, 6, size=3).tolist()
image = np.random.uniform(0.0, 1.0, size=tensor_shape + [1])
_, pseudo_inverse = matting.build_matrices(image, size=size)
a, b = matting.linear_coefficients(1.0 - image, pseudo_inverse)
reconstructed = matting.reconstruct(image, a, b)
self.assertAllClose(1.0 - image, reconstructed, atol=1e-4)
def b_fn(matte, pseudo_inverse):
_, b = matting.linear_coefficients(matte, pseudo_inverse)
return b
def a_fn(matte, pseudo_inverse):
a, _ = matting.linear_coefficients(matte, pseudo_inverse)
return a
