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
|
try:
from . import generic as g
except BaseException:
import generic as g
from io import BytesIO
from trimesh import voxel as v
from trimesh.exchange import binvox
from trimesh.voxel import runlength as rl
class BinvoxTest(g.unittest.TestCase):
def test_load_save_invariance(self):
np = g.np
n = 4
dense = np.random.uniform(size=(n,) * 3) > 0.8
dense[0, 0, 0] = dense[-1, -1, -1] = 1 # ensure extent test works
shape = dense.shape
rl_data = rl.dense_to_rle(dense.flatten(), dtype=np.uint8)
translate = np.array([2, 5, 10], dtype=np.float32)
scale = 5.0
base = binvox.voxel_from_binvox(
rl_data, shape, translate, scale, axis_order="xzy"
)
s = scale / (n - 1)
np.testing.assert_equal(
base.transform,
np.array([[s, 0, 0, 2], [0, s, 0, 5], [0, 0, s, 10], [0, 0, 0, 1]]),
)
dense = dense.transpose((0, 2, 1))
bound_min = translate - 0.5 * s
bound_max = translate + scale + 0.5 * s
np.testing.assert_allclose(base.bounds, [bound_min, bound_max])
np.testing.assert_equal(base.encoding.dense, dense)
if binvox.binvox_encoder is None:
g.log.warning("No binvox encoder found, skipping binvox export tests")
return
file_obj = BytesIO(binvox.export_binvox(base))
file_obj.seek(0)
loaded = binvox.load_binvox(file_obj)
np.testing.assert_equal(loaded.encoding.dense, base.encoding.dense)
self.assertTrue(isinstance(base, v.VoxelGrid))
self.assertTrue(isinstance(loaded, v.VoxelGrid))
np.testing.assert_equal(base.transform, loaded.transform)
np.testing.assert_equal(base.shape, loaded.shape)
if __name__ == "__main__":
g.trimesh.util.attach_to_log()
g.unittest.main()
|
