File size: 8,822 Bytes
36c95ba |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 |
import pytest
import torch
from torch.autograd import gradcheck
import kornia.geometry.epipolar as epi
from kornia.testing import assert_close
class TestIntrinsicsLike:
def test_smoke(self, device, dtype):
image = torch.rand(1, 3, 4, 4, device=device, dtype=dtype)
focal = torch.rand(1, device=device, dtype=dtype)
camera_matrix = epi.intrinsics_like(focal, image)
assert camera_matrix.shape == (1, 3, 3)
@pytest.mark.parametrize("batch_size", [1, 2, 4, 9])
def test_shape(self, batch_size, device, dtype):
B: int = batch_size
focal: float = 100.0
image = torch.rand(B, 3, 4, 4, device=device, dtype=dtype)
camera_matrix = epi.intrinsics_like(focal, image)
assert camera_matrix.shape == (B, 3, 3)
assert camera_matrix.device == image.device
assert camera_matrix.dtype == image.dtype
class TestScaleIntrinsics:
def test_smoke_float(self, device, dtype):
scale_factor: float = 1.0
camera_matrix = torch.rand(1, 3, 3, device=device, dtype=dtype)
camera_matrix_scale = epi.scale_intrinsics(camera_matrix, scale_factor)
assert camera_matrix_scale.shape == (1, 3, 3)
def test_smoke_tensor(self, device, dtype):
scale_factor = torch.tensor(1.0)
camera_matrix = torch.rand(1, 3, 3, device=device, dtype=dtype)
camera_matrix_scale = epi.scale_intrinsics(camera_matrix, scale_factor)
assert camera_matrix_scale.shape == (1, 3, 3)
@pytest.mark.parametrize("batch_size", [1, 2, 4, 9])
def test_shape(self, batch_size, device, dtype):
B: int = batch_size
scale_factor = torch.rand(B, device=device, dtype=dtype)
camera_matrix = torch.rand(B, 3, 3, device=device, dtype=dtype)
camera_matrix_scale = epi.scale_intrinsics(camera_matrix, scale_factor)
assert camera_matrix_scale.shape == (B, 3, 3)
def test_scale_double(self, device, dtype):
scale_factor = torch.tensor(0.5)
camera_matrix = torch.tensor(
[[[100.0, 0.0, 50.0], [0.0, 100.0, 50.0], [0.0, 0.0, 1.0]]], device=device, dtype=dtype
)
camera_matrix_expected = torch.tensor(
[[[50.0, 0.0, 25.0], [0.0, 50.0, 25.0], [0.0, 0.0, 1.0]]], device=device, dtype=dtype
)
camera_matrix_scale = epi.scale_intrinsics(camera_matrix, scale_factor)
assert_close(camera_matrix_scale, camera_matrix_expected, atol=1e-4, rtol=1e-4)
def test_gradcheck(self, device):
scale_factor = torch.ones(1, device=device, dtype=torch.float64, requires_grad=True)
camera_matrix = torch.ones(1, 3, 3, device=device, dtype=torch.float64)
assert gradcheck(epi.scale_intrinsics, (camera_matrix, scale_factor), raise_exception=True)
class TestProjectionFromKRt:
def test_smoke(self, device, dtype):
K = torch.rand(1, 3, 3, device=device, dtype=dtype)
R = torch.rand(1, 3, 3, device=device, dtype=dtype)
t = torch.rand(1, 3, 1, device=device, dtype=dtype)
P = epi.projection_from_KRt(K, R, t)
assert P.shape == (1, 3, 4)
@pytest.mark.parametrize("batch_size", [1, 2, 4])
def test_shape(self, batch_size, device, dtype):
B: int = batch_size
K = torch.rand(B, 3, 3, device=device, dtype=dtype)
R = torch.rand(B, 3, 3, device=device, dtype=dtype)
t = torch.rand(B, 3, 1, device=device, dtype=dtype)
P = epi.projection_from_KRt(K, R, t)
assert P.shape == (B, 3, 4)
def test_simple(self, device, dtype):
K = torch.tensor([[[10.0, 0.0, 30.0], [0.0, 20.0, 40.0], [0.0, 0.0, 1.0]]], device=device, dtype=dtype)
R = torch.tensor([[[1.0, 0.0, 0.0], [0.0, 1.0, 0.0], [0.0, 0.0, 1.0]]], device=device, dtype=dtype)
t = torch.tensor([[[1.0], [2.0], [3.0]]], device=device, dtype=dtype)
P_expected = torch.tensor(
[[[10.0, 0.0, 30.0, 100.0], [0.0, 20.0, 40.0, 160.0], [0.0, 0.0, 1.0, 3.0]]], device=device, dtype=dtype
)
P_estimated = epi.projection_from_KRt(K, R, t)
assert_close(P_estimated, P_expected, atol=1e-4, rtol=1e-4)
def test_krt_from_projection(self, device, dtype):
P = torch.tensor(
[[[10.0, 0.0, 30.0, 100.0], [0.0, 20.0, 40.0, 160.0], [0.0, 0.0, 1.0, 3.0]]], device=device, dtype=dtype
)
K_expected = torch.tensor([[[10.0, 0.0, 30.0], [0.0, 20.0, 40.0], [0.0, 0.0, 1.0]]], device=device, dtype=dtype)
R_expected = torch.tensor([[[1.0, 0.0, 0.0], [0.0, 1.0, 0.0], [0.0, 0.0, 1.0]]], device=device, dtype=dtype)
t_expected = torch.tensor([[[1.0], [2.0], [3.0]]], device=device, dtype=dtype)
K_estimated, R_estimated, t_estimated = epi.KRt_from_projection(P)
assert_close(K_estimated, K_expected, atol=1e-4, rtol=1e-4)
assert_close(R_estimated, R_expected, atol=1e-4, rtol=1e-4)
assert_close(t_estimated, t_expected, atol=1e-4, rtol=1e-4)
def test_gradcheck(self, device):
K = torch.rand(1, 3, 3, device=device, dtype=torch.float64, requires_grad=True)
R = torch.rand(1, 3, 3, device=device, dtype=torch.float64)
t = torch.rand(1, 3, 1, device=device, dtype=torch.float64)
assert gradcheck(epi.projection_from_KRt, (K, R, t), raise_exception=True)
class TestProjectionsFromFundamental:
def test_smoke(self, device, dtype):
F_mat = torch.rand(1, 3, 3, device=device, dtype=dtype)
P = epi.projections_from_fundamental(F_mat)
assert P.shape == (1, 3, 4, 2)
@pytest.mark.parametrize("batch_size", [1, 2, 4])
def test_shape(self, batch_size, device, dtype):
B: int = batch_size
F_mat = torch.rand(B, 3, 3, device=device, dtype=dtype)
P = epi.projections_from_fundamental(F_mat)
assert P.shape == (B, 3, 4, 2)
def test_gradcheck(self, device):
F_mat = torch.rand(1, 3, 3, device=device, dtype=torch.float64, requires_grad=True)
assert gradcheck(epi.projections_from_fundamental, (F_mat,), raise_exception=True)
class TestKRtFromProjection:
def test_smoke(self, device, dtype):
P = torch.randn(1, 3, 4, device=device, dtype=dtype)
K, R, t = epi.KRt_from_projection(P)
assert K.shape == (1, 3, 3)
assert R.shape == (1, 3, 3)
assert t.shape == (1, 3, 1)
@pytest.mark.parametrize("batch_size", [1, 2, 4])
def test_shape(self, batch_size, device, dtype):
B: int = batch_size
P = torch.rand(B, 3, 4, device=device, dtype=dtype)
K, R, t = epi.KRt_from_projection(P)
assert K.shape == (B, 3, 3)
assert R.shape == (B, 3, 3)
assert t.shape == (B, 3, 1)
def test_simple(self, device, dtype):
P = torch.tensor(
[[[308.0, 139.0, 231.0, 84.0], [481.0, 161.0, 358.0, 341.0], [384.0, 387.0, 459.0, 102.0]]],
device=device,
dtype=dtype,
)
K_expected = torch.tensor(
[[[17.006138, 122.441254, 390.211426], [0.0, 228.743622, 577.167480], [0.0, 0.0, 712.675232]]],
device=device,
dtype=dtype,
)
R_expected = torch.tensor(
[[[0.396559, 0.511023, -0.762625], [0.743249, -0.666318, -0.060006], [0.538815, 0.543024, 0.644052]]],
device=device,
dtype=dtype,
)
t_expected = torch.tensor([[[-6.477699], [1.129624], [0.143123]]], device=device, dtype=dtype)
K_estimated, R_estimated, t_estimated = epi.KRt_from_projection(P)
assert_close(K_estimated, K_expected, atol=1e-4, rtol=1e-4)
assert_close(R_estimated, R_expected, atol=1e-4, rtol=1e-4)
assert_close(t_estimated, t_expected, atol=1e-4, rtol=1e-4)
def test_projection_from_krt(self, device, dtype):
K = torch.tensor(
[[[17.006138, 122.441254, 390.211426], [0.0, 228.743622, 577.167480], [0.0, 0.0, 712.675232]]],
device=device,
dtype=dtype,
)
R = torch.tensor(
[[[0.396559, 0.511023, -0.762625], [0.743249, -0.666318, -0.060006], [0.538815, 0.543024, 0.644052]]],
device=device,
dtype=dtype,
)
t = torch.tensor([[[-6.477699], [1.129624], [0.143123]]], device=device, dtype=dtype)
P_expected = torch.tensor(
[[[308.0, 139.0, 231.0, 84.0], [481.0, 161.0, 358.0, 341.0], [384.0, 387.0, 459.0, 102.0]]],
device=device,
dtype=dtype,
)
P_estimated = epi.projection_from_KRt(K, R, t)
assert_close(P_estimated, P_expected, atol=1e-4, rtol=1e-4)
def test_gradcheck(self, device):
P_mat = torch.rand(1, 3, 4, device=device, dtype=torch.float64, requires_grad=True)
assert gradcheck(epi.KRt_from_projection, (P_mat,), raise_exception=True)
|