def test_export(self):
# the rig
rigs = kapture.Rigs()
rigs['rig0', get_camera_kapture_id_from_colmap_id(0)] = kapture.PoseTransform()
rigs['rig0', get_camera_kapture_id_from_colmap_id(1)] = kapture.PoseTransform()
# the records
records_camera = kapture.RecordsCamera()
records_camera[0, get_camera_kapture_id_from_colmap_id(0)] = 'camL/0000.jpg'
records_camera[0, get_camera_kapture_id_from_colmap_id(1)] = 'camR/0000.jpg'
records_camera[1, get_camera_kapture_id_from_colmap_id(0)] = 'camL/0001.jpg'
records_camera[1, get_camera_kapture_id_from_colmap_id(1)] = 'camR/0001.jpg'
# expect
expected_rigs = [{
"cameras": [
{
"camera_id": 0,
"image_prefix": "camL"
},
{
"camera_id": 1,
"image_prefix": "camR"
}
],
"ref_camera_id": 0
}]
colmap_camera_ids = {get_camera_kapture_id_from_colmap_id(i): i for i in range(2)}
colmap_rigs = export_colmap_rig_json(rigs, records_camera, colmap_camera_ids)
self.assertEqual(colmap_rigs, expected_rigs)
def test_trajectories_write(self):
pose1 = kapture.PoseTransform(r=[1.0, 0.0, 0.0, 0.0],
t=[0.0, 0.0, 0.0])
pose2 = kapture.PoseTransform(r=[0.5, 0.5, 0.5, 0.5], t=[4., 2., -2.])
content_expected = [
csv.KAPTURE_FORMAT_1 + kapture_linesep,
'# timestamp, device_id, qw, qx, qy, qz, tx, ty, tz' +
kapture_linesep,
' 0, cam0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0' +
kapture_linesep,
' 0, cam1, 0.5, 0.5, 0.5, 0.5, 4.0, 2.0, -2.0' +
kapture_linesep,
' 100, cam2, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0' +
kapture_linesep
]
trajectories = kapture.Trajectories()
timestamp1, timestamp2 = 0, 100
sensor_id1, sensor_id2, sensor_id3 = 'cam0', 'cam1', 'cam2'
trajectories[(timestamp1, sensor_id1)] = pose1
trajectories[(timestamp1, sensor_id2)] = pose2
trajectories[(timestamp2, sensor_id3)] = pose1
csv.trajectories_to_file(self._temp_filepath, trajectories)
with open(self._temp_filepath, 'rt') as f:
content_actual = f.readlines()
self.assertListEqual(content_actual, content_expected)
def setUp(self):
self._rigs = kapture.Rigs()
looking_not_straight = quaternion.from_rotation_vector(
[0, np.deg2rad(5.), 0])
self._rigs['rig', 'cam1'] = kapture.PoseTransform(
t=[-10, 0, 0], r=looking_not_straight).inverse()
self._rigs['rig', 'cam2'] = kapture.PoseTransform(
t=[+10, 0, 0], r=looking_not_straight.inverse()).inverse()
self._trajectories_rigs = kapture.Trajectories()
for timestamp, ratio in enumerate(np.linspace(0., 1., num=8)):
looking_around = quaternion.from_rotation_vector(
[0, np.deg2rad(360. * ratio), 0])
self._trajectories_rigs[timestamp, 'rig'] = kapture.PoseTransform(
t=[0, 0, -100.], r=looking_around)
self._trajectories_cams = kapture.Trajectories()
for timestamp, rig_id, pose_rig_from_world in kapture.flatten(
self._trajectories_rigs, is_sorted=True):
for rig_id2, cam_id, pose_cam_from_rig in kapture.flatten(
self._rigs):
assert rig_id == rig_id
pose_cam_from_world = kapture.PoseTransform.compose(
[pose_cam_from_rig, pose_rig_from_world])
self._trajectories_cams[timestamp,
cam_id] = pose_cam_from_world
def test_rig_remove(self):
rigs = kapture.Rigs()
rigs['rig0', 'cam0'] = kapture.PoseTransform(r=[1, 0, 0, 0],
t=[100, 0, 0])
rigs['rig0', 'cam1'] = kapture.PoseTransform(r=[1, 0, 0, 0],
t=[-100, 0, 0])
trajectories = kapture.Trajectories()
trajectories[0, 'rig0'] = kapture.PoseTransform(r=[1, 0, 0, 0],
t=[0, 0, 0])
trajectories[1, 'rig0'] = kapture.PoseTransform(r=[1, 0, 0, 0],
t=[0, 0, 10])
trajectories[2, 'rig0'] = kapture.PoseTransform(r=[1, 0, 0, 0],
t=[0, 0, 20])
trajectories_ = kapture.rigs_remove(trajectories, rigs)
# timestamps should be unchanged
self.assertEqual(trajectories_.keys(), trajectories.keys())
self.assertNotEqual(trajectories_.key_pairs(),
trajectories.key_pairs())
self.assertEqual(len(trajectories_.key_pairs()),
len(trajectories.key_pairs()) * len(rigs.key_pairs()))
self.assertIn((0, 'cam0'), trajectories_.key_pairs())
self.assertIn((0, 'cam1'), trajectories_.key_pairs())
self.assertIn((2, 'cam0'), trajectories_.key_pairs())
self.assertIn((2, 'cam1'), trajectories_.key_pairs())
self.assertAlmostEqual(trajectories_[2, 'cam1'].t_raw,
[-100.0, 0.0, 20.0])
self.assertAlmostEqual(trajectories_[2, 'cam1'].r_raw,
[1.0, 0.0, 0.0, 0.0])
def world_pose_transform_distance(pose_a: kapture.PoseTransform, pose_b: kapture.PoseTransform) -> Tuple[float, float]:
"""
get position and rotation error between two PoseTransform
pose_a and pose_b should be world to device
:return: (position_distance, rotation_distance in deg), can be nan is case of invalid comparison
"""
if pose_a.r is None and pose_a.t is None:
return math.nan, math.nan
elif pose_a.r is None:
device_to_world_a = kapture.PoseTransform(r=None, t=-pose_a.t)
elif pose_a.t is None:
device_to_world_a = kapture.PoseTransform(r=pose_a.r.inverse(), t=None)
else:
device_to_world_a = pose_a.inverse()
if pose_b.r is None and pose_b.t is None:
return math.nan, math.nan
elif pose_b.r is None:
device_to_world_b = kapture.PoseTransform(r=None, t=-pose_b.t)
elif pose_b.t is None:
device_to_world_b = kapture.PoseTransform(r=pose_b.r.inverse(), t=None)
else:
device_to_world_b = pose_b.inverse()
pose_error = pose_transform_distance(device_to_world_a, device_to_world_b)
return pose_error[0], np.rad2deg(pose_error[1])
def test_import(self):
colmap_rigs = [{
"cameras": [{
"camera_id": 0,
"image_prefix": "camL"
}, {
"camera_id": 1,
"image_prefix": "camR"
}],
"ref_camera_id":
0
}]
rigs_kapture, reconstructed_images, reconstructed_trajectories = import_colmap_rig_json(
rigs_colmap=colmap_rigs)
self.assertEqual([('rig0', get_camera_kapture_id_from_colmap_id(0)),
('rig0', get_camera_kapture_id_from_colmap_id(1))],
rigs_kapture.key_pairs())
self.assertIsNone(reconstructed_images)
self.assertIsNone(reconstructed_trajectories)
# the records
images = kapture.RecordsCamera()
images[0, get_camera_kapture_id_from_colmap_id(0)] = 'camL/0000.jpg'
images[1, get_camera_kapture_id_from_colmap_id(1)] = 'camR/0000.jpg'
images[2, get_camera_kapture_id_from_colmap_id(0)] = 'camL/0001.jpg'
images[3, get_camera_kapture_id_from_colmap_id(1)] = 'camR/0001.jpg'
rigs_kapture, reconstructed_images, reconstructed_trajectories = import_colmap_rig_json(
rigs_colmap=colmap_rigs, images=images)
# check timestamps has been recovered.
self.assertEqual([(0, get_camera_kapture_id_from_colmap_id(0)),
(0, get_camera_kapture_id_from_colmap_id(1)),
(1, get_camera_kapture_id_from_colmap_id(0)),
(1, get_camera_kapture_id_from_colmap_id(1))],
reconstructed_images.key_pairs())
# trajectories
trajectories = kapture.Trajectories()
trajectories[
0,
get_camera_kapture_id_from_colmap_id(0)] = kapture.PoseTransform()
trajectories[
1,
get_camera_kapture_id_from_colmap_id(1)] = kapture.PoseTransform()
trajectories[
2,
get_camera_kapture_id_from_colmap_id(0)] = kapture.PoseTransform()
trajectories[
3,
get_camera_kapture_id_from_colmap_id(1)] = kapture.PoseTransform()
rigs_kapture, reconstructed_images, reconstructed_trajectories = import_colmap_rig_json(
rigs_colmap=colmap_rigs, images=images, trajectories=trajectories)
self.assertEqual([(0, get_camera_kapture_id_from_colmap_id(0)),
(0, get_camera_kapture_id_from_colmap_id(1)),
(1, get_camera_kapture_id_from_colmap_id(0)),
(1, get_camera_kapture_id_from_colmap_id(1))],
reconstructed_trajectories.key_pairs())
def test_pose_write(self):
pose = kapture.PoseTransform(r=[1.0, 0.0, 0.0, 0.0], t=[0.0, 0.0, 0.0])
fields = csv.pose_to_list(pose)
self.assertListEqual(fields, [1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0])
pose = kapture.PoseTransform(r=[0.5, 0.5, 0.5, 0.5], t=[0., 0., 0.])
fields = csv.pose_to_list(pose)
self.assertListEqual(fields, [0.5, 0.5, 0.5, 0.5, 0.0, 0.0, 0.0])
pose = kapture.PoseTransform(r=[0.5, 0.5, 0.5, 0.5], t=[4., 2., -2.])
fields = csv.pose_to_list(pose)
self.assertListEqual(fields, [0.5, 0.5, 0.5, 0.5, 4., 2., -2.])
def setUp(self):
self.rotation_a = quaternion.quaternion(-0.572, 0.198, 0.755, -0.252)
self.rotation_a_negative = quaternion.quaternion(
0.572, -0.198, -0.755, 0.252)
self.translation_a = [144.801, -74.548, -17.746]
self.pose_a = kapture.PoseTransform(self.rotation_a,
self.translation_a)
self.pose_a_negative = kapture.PoseTransform(self.rotation_a_negative,
self.translation_a)
self.rotation_b = quaternion.quaternion(0.878, 0.090, 0.374, -0.285)
self.rotation_b_negative = quaternion.quaternion(
-0.878, -0.090, -0.374, 0.285)
self.translation_b = [4.508, 45.032, -37.840]
self.pose_b = kapture.PoseTransform(self.rotation_b,
self.translation_b)
self.pose_b_negative = kapture.PoseTransform(self.rotation_b_negative,
self.translation_b)
self.pose_ab = kapture.PoseTransform(self.rotation_a,
self.translation_b)
self.pose_ba = kapture.PoseTransform(self.rotation_b,
self.translation_a)
self.pose_none = kapture.PoseTransform(r=None, t=None)
self.pose_r_none = kapture.PoseTransform(r=None, t=self.translation_a)
self.pose_t_none = kapture.PoseTransform(r=self.rotation_a, t=None)
def test_timestamps_length(self):
trajectories = kapture.Trajectories()
self.assertEqual(trajectories.timestamp_length(), -1)
trajectories[1614362592378, 'cam0'] = kapture.PoseTransform(r=[1, 0, 0, 0], t=[0, 0, 20])
trajectories[1614362592634, 'cam0'] = kapture.PoseTransform(r=[1, 0, 0, 0], t=[0, 0, 0])
trajectories[1614362592378, 'cam1'] = kapture.PoseTransform(r=[1, 0, 0, 0], t=[10, 0, 20])
trajectories[1614362593123, 'cam1'] = kapture.PoseTransform(r=[1, 0, 0, 0], t=[10, 0, 30])
self.assertEqual(trajectories.timestamp_length(), 13)
# Check that if we have timestamps of different precision, we can not compute a common length
trajectories[1614362594, 'lidar0'] = kapture.PoseTransform(r=[1, 0, 0, 0], t=[0, 0, 0])
self.assertEqual(trajectories.timestamp_length(), -1)
def test_identity(self):
# default construct
pose_identity = kapture.PoseTransform()
actual_points3d = pose_identity.transform_points(
self.expected_points3d)
self.assertTrue(np.all(actual_points3d == self.expected_points3d))
# explicit identity
pose_identity = kapture.PoseTransform(r=[1, 0, 0, 0], t=[0, 0, 0])
actual_points3d = pose_identity.transform_points(
self.expected_points3d)
self.assertTrue(np.all(actual_points3d == self.expected_points3d))
def test_init(self):
rigs = kapture.Rigs()
rigs['rig0', 'cam0'] = kapture.PoseTransform()
rigs['rig0', 'cam1'] = kapture.PoseTransform()
rigs['rig1'] = {
'cam2': kapture.PoseTransform(),
'cam3': kapture.PoseTransform()
}
self.assertEqual(2, len(rigs))
self.assertIn('rig0', rigs)
self.assertIn('rig1', rigs)
self.assertIn('cam0', rigs['rig0'])
self.assertNotIn('cam0', rigs['rig1'])
def test_rig_write(self):
rigs = kapture.Rigs()
rigs['rig1', 'cam0'] = kapture.PoseTransform()
rigs['rig1', 'cam1'] = kapture.PoseTransform(r=[0.5, 0.5, 0.5, 0.5])
content_expected = '\n'.join([csv.KAPTURE_FORMAT_1,
'# rig_id, sensor_id, qw, qx, qy, qz, tx, ty, tz',
'rig1, cam0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0',
'rig1, cam1, 0.5, 0.5, 0.5, 0.5, 0.0, 0.0, 0.0',
''])
csv.rigs_to_file(self._temp_filepath, rigs)
with open(self._temp_filepath, 'rt') as f:
content_actual = ''.join(f.readlines())
self.assertEqual(content_expected, content_actual)
def _make_rigs(replace_pose_rig, trajectories) -> kapture.Rigs:
logger.info('Making up a rig ...')
rigs = kapture.Rigs()
pose_babord = kapture.PoseTransform(t=[0, 0, 0],
r=quaternion.from_rotation_vector(
[0, -np.pi / 2, 0]))
pose_tribord = kapture.PoseTransform(t=[0, 0, 0],
r=quaternion.from_rotation_vector(
[0, np.pi / 2, 0]))
rigs['silda_rig', '0'] = pose_babord
rigs['silda_rig', '1'] = pose_tribord
if replace_pose_rig:
logger.info('replacing camera poses with rig poses.')
kapture.rigs_recover_inplace(trajectories, rigs)
return rigs
def setUp(self):
self._samples_folder = path.abspath(
path.join(path.dirname(__file__), '..', 'samples', 'm1x'))
self._kapture_data = csv.kapture_from_dir(self._samples_folder)
rotation_a = quaternion.quaternion(-0.572, 0.198, 0.755, -0.252)
rotation_b = quaternion.quaternion(0.878, 0.090, 0.374, -0.285)
translation_a = [144.801, -74.548, -17.746]
translation_b = [144.701, -73.548, -17.746]
self._pose_a = kapture.PoseTransform(rotation_a, translation_a)
self._pose_b = kapture.PoseTransform(rotation_b, translation_b)
self._rigs = kapture.Rigs()
self._rigs['rig_a', '144323'] = self._pose_a
self._rigs['rig_b', '144323'] = self._pose_b
def test_translation(self):
pose_translate = kapture.PoseTransform(r=[1, 0, 0, 0], t=[0, 0, 10])
actual_points3d = pose_translate.transform_points(
self.expected_points3d)
diff = (actual_points3d - self.expected_points3d)
self.assertTrue(np.all(diff[:, 2] == 10))
self.assertTrue(np.all(diff[:, 0:2] == 0))
def test_trajectories_read(self):
content = [
'# timestamp, device_id, qw, qx, qy, qz, tx, ty, tz',
'0, cam0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0',
' 0, cam1, 0.5, 0.5, 0.5, 0.5, 4.0, 2.0, -2.0',
' 100, cam2, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0'
]
with open(self._temp_filepath, 'wt') as f:
f.write(kapture_linesep.join(content))
device_ids = {'cam0', 'cam1', 'cam2'}
trajectories = csv.trajectories_from_file(self._temp_filepath,
device_ids)
self.assertIsInstance(trajectories, kapture.Trajectories)
self.assertEqual(2, len(trajectories.keys())) # timestamps
self.assertEqual(3,
len(trajectories.key_pairs())) # timestamp x devices
self.assertIn(0, trajectories)
self.assertIn(100, trajectories)
self.assertIn('cam0', trajectories[0])
self.assertIn('cam1', trajectories[0])
pose = kapture.PoseTransform(r=[0.5, 0.5, 0.5, 0.5], t=[4., 2., -2.])
self.assertAlmostEqual(trajectories[(0, 'cam1')].r_raw, pose.r_raw)
self.assertAlmostEqual(trajectories[(0, 'cam1')].t_raw, pose.t_raw)
def rigs_from_file(filepath: str, sensor_ids: Set[str]) -> kapture.Rigs:
"""
Reads rigs from CSV file.
:param filepath: input file path
:param sensor_ids: input set of valid sensor ids.
If a rig id collides one of them, raise error.
If a sensor in rig is not in sensor_ids, it is ignored.
:return: rigs
"""
# rig_id, sensor_id, qw, qx, qy, qz, tx, ty, tz
rigs = kapture.Rigs()
with open(filepath) as file:
table = table_from_file(file)
for rig_id, sensor_id, qw, qx, qy, qz, tx, ty, tz in table:
if rig_id in sensor_ids:
raise ValueError(f'collision between a sensor ID and rig ID ({rig_id})')
if sensor_id not in sensor_ids:
# just ignore
continue
rotation = float_array_or_none([qw, qx, qy, qz])
translation = float_array_or_none([tx, ty, tz])
pose = kapture.PoseTransform(rotation, translation)
# rigs.setdefault(str(rig_id), kapture.Rig())[sensor_id] = pose
rigs[str(rig_id), sensor_id] = pose
return rigs
def test_t265_db_only(self):
kapture_data = import_colmap_database(self._database_filepath,
self._kapture_dirpath,
no_geometric_filtering=True)
# check the numbers
self.assertEqual(2, len(kapture_data.sensors))
self.assertEqual(6, len(kapture_data.trajectories))
self.assertEqual(6, len(kapture_data.records_camera))
# check camera ids
camera_ids_expected = set(['cam_00001', 'cam_00002'
]) # may evolve in future, not crucial
camera_ids_actual = set(kapture_data.sensors.keys())
self.assertEqual(camera_ids_expected, camera_ids_actual)
# check camera ids consistent in trajectories
camera_ids_trajectories = set(
cam_id
for _, cam_id, _ in kapture.flatten(kapture_data.trajectories))
self.assertEqual(camera_ids_actual, camera_ids_trajectories)
# check camera ids consistent in records_camera
camera_ids_records = set(
cam_id
for _, cam_id, _ in kapture.flatten(kapture_data.records_camera))
self.assertEqual(camera_ids_actual, camera_ids_records)
# check camera parameters
cam1 = kapture_data.sensors['cam_00001']
self.assertIsInstance(cam1, kapture.Camera)
self.assertEqual('camera', cam1.sensor_type)
self.assertEqual(kapture.CameraType.OPENCV_FISHEYE, cam1.camera_type)
params_expected = [
848.0, 800.0, 284.468, 285.51, 424.355, 393.742, 0.0008, 0.031,
-0.03, 0.005
]
self.assertAlmostEqual(params_expected, cam1.camera_params)
# check records
timestamp, cam_id, image = next(
kapture.flatten(kapture_data.records_camera, is_sorted=True))
self.assertEqual(1, timestamp)
self.assertEqual('cam_00002', cam_id)
self.assertEqual('rightraw/frame_000000001.jpg', image)
# check trajectories
timestamp, cam_id, pose = next(
kapture.flatten(kapture_data.trajectories, is_sorted=True))
self.assertEqual(1, timestamp)
self.assertEqual('cam_00002', cam_id)
pose_expected = kapture.PoseTransform(
r=[
0.9540331248716523, -0.03768128483784883, -0.2972570621910482,
-0.0062565444214723875
],
t=[2.7109402281860904, 0.13236653865769618, -2.868626176500939])
self.assertTrue(equal_poses(pose_expected, pose))
# this sample has no keypoints, descriptors nor matches
self.assertFalse(
path.exists(path.join(self._kapture_dirpath, 'reconstruction')))
def rig_to_ply_stream(stream, rig: Dict[str, kapture.PoseTransform], axis_length: float = 1.) -> None:
"""
Writes the rig to a stream.
:param stream: an open stream to write to
:param rig: rig to write
:param axis_length: length of the axis
"""
device_list = [(cam_id, pose_tr)
for cam_id, pose_tr in rig.items()]
# add the rig center in devices
device_list = [(-1, kapture.PoseTransform())] + device_list
# create 4 points per device: 1 for center, 3 for axis
points_colored_list = []
edges_list = []
for cam_id, pose_tr in device_list:
axis = get_axis_in_world(pose_tr, axis_length)
edges_list += [(0, len(points_colored_list))]
points_colored_list += [p + AXIS_COLORS[i] for i, p in enumerate(axis.tolist())]
# write points into ply
header_to_ply_stream(stream,
nb_vertex=len(points_colored_list),
nb_edges=len(edges_list))
for p3d in points_colored_list:
line = ['{:<25}'.format(i) for i in p3d[0:3]]
line += ['{:03}'.format(int(i)) for i in p3d[3:6]]
stream.write(' '.join(line) + '\n')
for e in edges_list:
line = ['{:2}'.format(i) for i in e]
stream.write(' '.join(line) + '\n')
def read_robotcar_v2_train(
robotcar_path: str) -> Dict[str, kapture.PoseTransform]:
"""
Read the robot car v2 train data file
:param robotcar_path: path to the data file
:return: train data
"""
with (open(path.join(robotcar_path, 'robotcar_v2_train.txt'), 'r')) as f:
lines = f.readlines()
# remove empty lines
lines = (l2 for l2 in lines if l2.strip())
# trim trailing EOL
lines = (l3.rstrip("\n\r") for l3 in lines)
# split space
lines = (l4.split() for l4 in lines)
lines = list(lines)
train_data = {
table[0].replace(".png", ".jpg"): kapture.PoseTransform(
r=quaternion.from_rotation_matrix(
[[float(v) for v in table[1:4]],
[float(v) for v in table[5:8]],
[float(v) for v in table[9:12]]]),
t=[float(v)
for v in [table[4], table[8], table[12]]]).inverse()
for table in lines
}
return train_data
def _import_trajectories(silda_dir_path, image_name_to_ids,
hide_progress_bars) -> kapture.Trajectories:
logger.info('Processing trajectories ...')
trajectories = kapture.Trajectories()
with open(path.join(silda_dir_path, 'silda-train-poses.txt')) as file:
lines = file.readlines()
lines = (line.rstrip().split() for line in lines)
extrinsics = {
line[0]: np.array(line[1:8], dtype=np.float)
for line in lines
}
for silda_image_name, pose_params in tqdm(extrinsics.items(),
disable=hide_progress_bars):
# Silda poses are 7-dim vectors with the rotation quaternion,
# and the translation vector. The order needs to be:
# qw,qx,qy,qz,tx,ty,tz
# The parameters should be described in terms of camera to world transformations
if silda_image_name not in image_name_to_ids:
# if this is not referenced: means its part of the corpus to be ignored.
continue
pose = kapture.PoseTransform(pose_params[0:4],
pose_params[4:7]).inverse()
timestamp, cam_id = image_name_to_ids[silda_image_name]
trajectories[timestamp, cam_id] = pose
# if query, trajectories is empty, so juste do not save it
if len(trajectories) == 0:
trajectories = None
return trajectories
def average_pose_transform_weighted(poses: List[kapture.PoseTransform],
weights: Union[List[float], np.ndarray]) -> kapture.PoseTransform:
"""
average a list of poses with any weights
:param poses: list of poses to average
:param weights: a (len(poses),) vector
:return: average PoseTransform
"""
assert isinstance(poses, list)
assert len(poses) > 0
assert isinstance(weights, list)
assert len(weights) == len(poses)
weight_sum = np.sum(weights)
# handle NoneType with try expect blocks
try:
translation = np.sum(tuple([weights[i] * pose.t for i, pose in enumerate(poses)]), axis=0) / weight_sum
except TypeError:
translation = None
try:
rotation = average_quaternion_weighted(np.vstack(tuple([quaternion.as_float_array(pose.r) for pose in poses])),
weights)
except TypeError:
rotation = None
return kapture.PoseTransform(r=rotation, t=translation)
def convert_testing_extrinsics(offset: int,
testing_extrinsics: Iterable[VirtualGalleryTestingExtrinsic],
images: kapture.RecordsCamera,
trajectories: kapture.Trajectories) -> None:
"""
Import all testing extrinsics into the images and trajectories.
:param offset:
:param testing_extrinsics: testing extrinsics to import
:param images: image list to add to
:param trajectories: trajectories to add to
"""
# Map (light_id, loop_id, frame_id) to a unique timestamp
testing_frames_tuples = ((extrinsic.light_id, extrinsic.occlusion_id, extrinsic.frame_id)
for extrinsic in testing_extrinsics)
testing_frames_tuples = OrderedDict.fromkeys(testing_frames_tuples).keys()
testing_frame_mapping = {v: n + offset for n, v in enumerate(testing_frames_tuples)}
# Export images and trajectories
logger.info("Converting testing images and trajectories...")
for extrinsic in testing_extrinsics:
rotation_matrix = [extrinsic.extrinsics[0:3], extrinsic.extrinsics[4:7], extrinsic.extrinsics[8:11]]
rotation = quaternion.from_rotation_matrix(rotation_matrix)
timestamp = testing_frame_mapping[(extrinsic.light_id, extrinsic.occlusion_id, extrinsic.frame_id)]
camera_device_id = _get_testing_camera_name(extrinsic.light_id, extrinsic.occlusion_id, extrinsic.frame_id)
translation_vector = [extrinsic.extrinsics[3], extrinsic.extrinsics[7], extrinsic.extrinsics[11]]
images[(timestamp, camera_device_id)] = (f'testing/gallery_light{extrinsic.light_id}_occlusion'
f'{extrinsic.occlusion_id}/frames/rgb/'
f'camera_0/rgb_{extrinsic.frame_id:05}.jpg')
trajectories[(timestamp, camera_device_id)] = kapture.PoseTransform(rotation, translation_vector)
def test_type_checking(self):
records_camera = kapture.RecordsCamera()
valid_ts, valid_id, valid_record = 0, 'cam0', 'cam0/image0.jpg'
invalid_ts, invalid_id, invalid_record = '0', float(
0), kapture.PoseTransform()
self.assertRaises(TypeError, records_camera.__setitem__,
(invalid_ts, valid_id), valid_record)
self.assertRaises(TypeError, records_camera.__setitem__,
(valid_ts, invalid_id), valid_record)
self.assertRaises(TypeError, records_camera.__setitem__,
(valid_ts, valid_id), invalid_record)
self.assertRaises(TypeError, records_camera.__setitem__,
(invalid_ts, invalid_id), invalid_record)
self.assertRaises(TypeError, records_camera.__setitem__, invalid_ts,
{valid_id: valid_record})
self.assertRaises(TypeError, records_camera.__setitem__, valid_ts,
{invalid_id: valid_record})
self.assertRaises(TypeError, records_camera.__setitem__, valid_ts,
{valid_id: invalid_record})
self.assertRaises(TypeError, records_camera.__setitem__, invalid_ts,
valid_record)
self.assertRaises(TypeError, records_camera.__contains__, invalid_ts,
valid_id)
self.assertRaises(TypeError, records_camera.__contains__, valid_ts,
invalid_id)
self.assertRaises(TypeError, records_camera.__contains__, invalid_ts,
invalid_id)
self.assertRaises(TypeError, records_camera.__delitem__, invalid_ts)
self.assertRaises(TypeError, records_camera.__delitem__,
(valid_ts, invalid_id))
def test_type_checking(self):
traj = kapture.Trajectories()
valid_ts, valid_id, valid_pose = 0, 'cam0', kapture.PoseTransform()
invalid_ts, invalid_id, invalid_pose = '0', float(0), 'pose'
self.assertRaises(TypeError, traj.__setitem__, (invalid_ts, valid_id),
valid_pose)
self.assertRaises(TypeError, traj.__setitem__, (valid_ts, invalid_id),
valid_pose)
self.assertRaises(TypeError, traj.__setitem__, (valid_ts, valid_id),
invalid_pose)
self.assertRaises(TypeError, traj.__setitem__,
(invalid_ts, invalid_id), invalid_pose)
self.assertRaises(TypeError, traj.__setitem__, invalid_ts,
{valid_id: valid_pose})
self.assertRaises(TypeError, traj.__setitem__, valid_ts,
{invalid_id: valid_pose})
self.assertRaises(TypeError, traj.__setitem__, valid_ts,
{valid_id: invalid_pose})
self.assertRaises(TypeError, traj.__setitem__, invalid_ts, valid_pose)
self.assertRaises(TypeError, traj.__contains__, invalid_ts, valid_id)
self.assertRaises(TypeError, traj.__contains__, valid_ts, invalid_id)
self.assertRaises(TypeError, traj.__contains__, invalid_ts, invalid_id)
self.assertRaises(TypeError, traj.__delitem__, invalid_ts)
self.assertRaises(TypeError, traj.__delitem__, (valid_ts, invalid_id))
def import_robotcar_rig(extrinsics_dir_path: str) -> kapture.Rigs:
"""
Read extrinsics files and convert to kapture rigs
:param extrinsics_dir_path: path to directory with extrinsics
:return: kapture.Rigs object
"""
# From dataset documentation:
# The extrinsic parameters of the three cameras on the car with respect to the car itself.
# The extrinsics are provided as a 4x4 matrix
# [R c]
# [0 1]
# where R is a rotation matrix that maps from camera to world coordinates and c is the position of the camera in
# world coordinates. The entries of the matrix are separated by ,.
rigs = kapture.Rigs()
for root, dirs, files in os.walk(extrinsics_dir_path):
for extrinsics_filepath in files:
(camera_id, _) = extrinsics_filepath.split('_')
extrinsic_file = open(
path.join(extrinsics_dir_path, extrinsics_filepath), 'r')
with extrinsic_file as f:
m = np.array([[float(num) for num in line.split(',')]
for line in f])
rotation_matrix = m[0:3, 0:3]
position = m[0:3, 3:4]
pose_transform = kapture.PoseTransform(
t=position, r=quaternion.from_rotation_matrix(rotation_matrix))
# kapture rig format is "rig to sensor"
rigs['car', camera_id] = pose_transform.inverse()
return rigs
def _verify_data(self, kapture_data) -> None:
cameras = kapture_data.cameras
self.assertIsNotNone(cameras, "Cameras exist")
self.assertEqual(1, len(cameras), "One camera")
camera = next(iter(
cameras.values())) # just take the first camera defined
self.assertEqual(camera.camera_type,
kapture.CameraType.SIMPLE_RADIAL_FISHEYE,
"Type fisheye")
camera_params = camera.camera_params
self.assertEqual(848, camera_params[0], "width")
self.assertEqual(800, camera_params[1], "height")
records_camera = kapture_data.records_camera
self.assertEqual(5, len(records_camera), "Number of images")
first_record = records_camera[0]
img_path = next(iter(first_record.values()))
self.assertEqual("images/frame_000000001.jpg", img_path, "Image path")
trajectories = kapture_data.trajectories
self.assertEqual(5, len(trajectories), "Trajectories points")
k_pose6d = next(iter(
trajectories[0].values())) # Kapture.PoseTransform
ref_pose = kapture.PoseTransform(t=FIRST_TRAJECTORY_TRANSLATION,
r=FIRST_TRAJECTORY_ROTATION)
self.assertTrue(equal_poses(ref_pose, k_pose6d),
"First trajectory pose")
self.assertIsNone(kapture_data.keypoints, "No keypoints")
self.assertIsNone(kapture_data.observations, "No observations")
self.assertIsNone(kapture_data.points3d, "No 3D points")
def convert_results_to_kapture(query_path: str, results: str, outpath: str):
"""
convert file with name qw qx qy qz tx ty tz to kapture
"""
skip_heavy_useless = [
kapture.Trajectories, kapture.RecordsLidar, kapture.RecordsWifi,
kapture.Keypoints, kapture.Descriptors, kapture.GlobalFeatures,
kapture.Matches, kapture.Points3d, kapture.Observations
]
kapture_query = kapture_from_dir(query_path, skip_list=skip_heavy_useless)
inverse_records_camera = {
image_name: (timestamp, sensor_id)
for timestamp, sensor_id, image_name in kapture.flatten(
kapture_query.records_camera)
}
trajectories = kapture.Trajectories()
with open(results) as fid:
lines = fid.readlines()
lines = [line.rstrip().split() for line in lines if line != '\n']
for line in lines:
image_name = line[0]
rotation = quaternion.quaternion(float(line[1]), float(line[2]),
float(line[3]), float(line[4]))
translation = [float(line[5]), float(line[6]), float(line[7])]
timestamp, sensor_id = inverse_records_camera[image_name]
trajectories[timestamp,
sensor_id] = kapture.PoseTransform(rotation, translation)
kapture_query.trajectories = trajectories
kapture_to_dir(outpath, kapture_query)
def convert_training_extrinsics(
offset: int,
training_extrinsics: Iterable[VirtualGalleryTrainingExtrinsic],
images: kapture.RecordsCamera, trajectories: kapture.Trajectories,
as_rig: bool) -> None:
"""
Import all training extrinsics into the images and trajectories.
:param offset:
:param training_extrinsics: training extrinsics to import
:param images: image list to add to
:param trajectories: trajectories to add to
:param as_rig: writes the position of the rig instead of individual cameras
"""
# Map (light_id, loop_id, frame_id) to a unique timestamp
training_frames_tuples = ((extrinsic.light_id, extrinsic.loop_id,
extrinsic.frame_id)
for extrinsic in training_extrinsics)
training_frames_tuples = OrderedDict.fromkeys(
training_frames_tuples).keys()
training_frame_mapping = {
v: n + offset
for n, v in enumerate(training_frames_tuples)
}
# Export images and trajectories
logger.info("Converting training images and trajectories...")
for extrinsic in training_extrinsics:
rotation_matrix = [
extrinsic.extrinsics[0:3], extrinsic.extrinsics[4:7],
extrinsic.extrinsics[8:11]
]
rotation = quaternion.from_rotation_matrix(rotation_matrix)
timestamp = training_frame_mapping[(extrinsic.light_id,
extrinsic.loop_id,
extrinsic.frame_id)]
camera_device_id = _get_training_camera_name(extrinsic.camera_id)
translation_vector = [
extrinsic.extrinsics[3], extrinsic.extrinsics[7],
extrinsic.extrinsics[11]
]
images[(timestamp, camera_device_id)] = (
f'training/gallery_light{extrinsic.light_id}_loop'
f'{extrinsic.loop_id}/frames/rgb/camera_{extrinsic.camera_id}'
f'/rgb_{extrinsic.frame_id:05}.jpg')
pose_cam = kapture.PoseTransform(rotation, translation_vector)
if as_rig:
pose_rig = kapture.PoseTransform.compose([
training_rig_config[(_get_training_rig_name(),
camera_device_id)].inverse(), pose_cam
])
if (timestamp, _get_training_rig_name()) in trajectories:
assert equal_poses(
pose_rig,
trajectories[(timestamp, _get_training_rig_name())])
else:
trajectories[(timestamp, _get_training_rig_name())] = pose_rig
else:
trajectories[(timestamp, camera_device_id)] = pose_cam
def test_none(self):
expect_r = None
expect_t = None
pose = kapture.PoseTransform(expect_r, expect_t)
actual_r = pose.r_raw
actual_t = pose.t_raw
self.assertEqual(actual_r, None)
self.assertEqual(actual_t, None)