-
Notifications
You must be signed in to change notification settings - Fork 3
/
data_utils.py
executable file
·133 lines (110 loc) · 4.17 KB
/
data_utils.py
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
import numpy as np
def normal_pc(pc):
"""
normalize point cloud in range L
:param pc: type list
:return: type list
"""
pc_mean = pc.mean(axis=0)
pc = pc - pc_mean
pc_L_max = np.max(np.sqrt(np.sum(abs(pc ** 2), axis=-1)))
pc = pc/pc_L_max
return pc
def rotation_point_cloud(pc):
"""
Randomly rotate the point clouds to augment the dataset
rotation is per shape based along up direction
:param pc: B X N X 3 array, original batch of point clouds
:return: BxNx3 array, rotated batch of point clouds
"""
# rotated_data = np.zeros(pc.shape, dtype=np.float32)
rotation_angle = np.random.uniform() * 2 * np.pi
cosval = np.cos(rotation_angle)
sinval = np.sin(rotation_angle)
rotation_matrix = np.array([[cosval, 0, sinval],
[0, 1, 0],
[-sinval, 0, cosval]])
rotated_data = np.dot(pc.reshape((-1, 3)), rotation_matrix)
return rotated_data
def rotate_point_cloud_by_angle(pc, rotation_angle):
"""
Randomly rotate the point clouds to augment the dataset
rotation is per shape based along up direction
:param pc: B X N X 3 array, original batch of point clouds
:param rotation_angle: angle of rotation
:return: BxNx3 array, rotated batch of point clouds
"""
# rotated_data = np.zeros(pc.shape, dtype=np.float32)
# rotation_angle = np.random.uniform() * 2 * np.pi
cosval = np.cos(rotation_angle)
sinval = np.sin(rotation_angle)
rotation_matrix = np.array([[cosval, 0, sinval],
[0, 1, 0],
[-sinval, 0, cosval]])
rotated_data = np.dot(pc.reshape((-1, 3)), rotation_matrix)
return rotated_data
def jitter_point_cloud(pc, sigma=0.01, clip=0.05):
"""
Randomly jitter points. jittering is per point.
:param pc: B X N X 3 array, original batch of point clouds
:param sigma:
:param clip:
:return:
"""
jittered_data = np.clip(sigma * np.random.randn(*pc.shape), -1 * clip, clip)
jittered_data += pc
return jittered_data
def shift_point_cloud(pc, shift_range=0.1):
""" Randomly shift point cloud. Shift is per point cloud.
Input:
BxNx3 array, original batch of point clouds
Return:
BxNx3 array, shifted batch of point clouds
"""
N, C = pc.shape
shifts = np.random.uniform(-shift_range, shift_range, 3)
pc += shifts
return pc
def random_scale_point_cloud(pc, scale_low=0.8, scale_high=1.25):
""" Randomly scale the point cloud. Scale is per point cloud.
Input:
BxNx3 array, original batch of point clouds
Return:
BxNx3 array, scaled batch of point clouds
"""
N, C = pc.shape
scales = np.random.uniform(scale_low, scale_high, 1)
pc *= scales
return pc
def rotate_perturbation_point_cloud(pc, angle_sigma=0.06, angle_clip=0.18):
""" Randomly perturb the point clouds by small rotations
Input:
BxNx3 array, original batch of point clouds
Return:
BxNx3 array, rotated batch of point clouds
"""
# rotated_data = np.zeros(pc.shape, dtype=np.float32)
angles = np.clip(angle_sigma * np.random.randn(3), -angle_clip, angle_clip)
Rx = np.array([[1, 0, 0],
[0, np.cos(angles[0]), -np.sin(angles[0])],
[0, np.sin(angles[0]), np.cos(angles[0])]])
Ry = np.array([[np.cos(angles[1]), 0, np.sin(angles[1])],
[0, 1, 0],
[-np.sin(angles[1]), 0, np.cos(angles[1])]])
Rz = np.array([[np.cos(angles[2]), -np.sin(angles[2]), 0],
[np.sin(angles[2]), np.cos(angles[2]), 0],
[0, 0, 1]])
R = np.dot(Rz, np.dot(Ry, Rx))
shape_pc = pc
rotated_data = np.dot(shape_pc.reshape((-1, 3)), R)
return rotated_data
def pc_augment(pc, angle):
pc = rotate_point_cloud_by_angle(pc, angle)
# pc = rotation_point_cloud(pc)
# pc = jitter_point_cloud(pc)
# pc = random_scale_point_cloud(pc)
# pc = rotate_perturbation_point_cloud(pc)
# pc = shift_point_cloud(pc)
return pc
def data_augment(train_data, angle):
return pc_augment(train_data, angle).reshape(-1, train_data.shape[1], train_data.shape[2])