-
Notifications
You must be signed in to change notification settings - Fork 28
/
Copy pathVisualization.py
258 lines (194 loc) · 9.78 KB
/
Visualization.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
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
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Mon Jul 1 10:27:44 2019
@author: rain
"""
import numpy as np
import mayavi.mlab as mlab
import matplotlib.pyplot as plt
from numpy import linalg as LA
from numpy import dot as dot
from Transformations import *
def ShowTrajactory(fig, poses, color, ptSize):
Rs, Ts = GetRsAndTsFromPoses(poses)
AxisLength = 0.8
Rs_ = Rs*AxisLength;
colors = np.ones((Ts.shape[0],),dtype=np.float32)*color
if ptSize == 0:
node = mlab.points3d(Ts[:,0], Ts[:,1], Ts[:,2], mode="point", figure=fig)
else:
node = mlab.points3d(Ts[:,0], Ts[:,1], Ts[:,2], scale_factor=ptSize, figure=fig)
node.glyph.scale_mode = 'scale_by_vector'
node.mlab_source.dataset.point_data.scalars = colors
for iAxis in range(3):
node = mlab.quiver3d(Ts[:,0], Ts[:,1], Ts[:,2], \
Rs_[:,iAxis,0], Rs_[:,iAxis,1], Rs_[:,iAxis,2], \
figure=fig, line_width=2.0, scale_factor=1, mode='2ddash')
def CompareTrajactory(fig, trajectory0, trajectory1, nStartFrame, nEndFrame):
if nEndFrame <= 0:
nEndFrame = trajectory0.shape[0]
node = mlab.quiver3d(trajectory0[nStartFrame:nEndFrame,0], trajectory0[nStartFrame:nEndFrame,1], trajectory0[nStartFrame:nEndFrame,2], \
trajectory1[nStartFrame:nEndFrame,0]-trajectory0[nStartFrame:nEndFrame,0], \
trajectory1[nStartFrame:nEndFrame,1]-trajectory0[nStartFrame:nEndFrame,1], \
trajectory1[nStartFrame:nEndFrame,2]-trajectory0[nStartFrame:nEndFrame,2], \
figure=fig, line_width=0.5, scale_factor=1, mode='2ddash')
# mlab.axes(x_axis_visibility = True)
# mlab.axes(y_axis_visibility = True)
# mlab.axes(z_axis_visibility = True)
def ShowMatchingResult(RawDataDir, iFrame0, iFrame1, KeyPts0, KeyPts1, PlanarPts0, PlanarPts1,
bShowPairs, inliersIdx0, inliersIdx1, oriRelR, oriRelT, deltaR, deltaT):
rawFile0 = RawDataDir + str(iFrame0).zfill(6)+'.bin'
rawFile1 = RawDataDir + str(iFrame1).zfill(6)+'.bin'
PC0 = np.fromfile(rawFile0, dtype=np.float32, count=-1).reshape([-1,4])[:,0:3]
PC1 = np.fromfile(rawFile1, dtype=np.float32, count=-1).reshape([-1,4])[:,0:3]
pairs0 = KeyPts0[inliersIdx0,:]
pairs1 = KeyPts1[inliersIdx1,:]
PC1_ = (np.dot(oriRelR, PC1.T) + oriRelT.reshape(3,1)).T
KeyPts1_ = (np.dot(oriRelR, KeyPts1.T) + oriRelT.reshape(3,1)).T
pairs1_ = (np.dot(oriRelR, pairs1.T) + oriRelT.reshape(3,1)).T
PlanarPts1_ = (np.dot(oriRelR, PlanarPts1.T) + oriRelT.reshape(3,1)).T
PC1__ = (np.dot(deltaR, PC1_.T) + deltaT.reshape(3,1)).T
FusedPC__ = np.r_[PC0, PC1__]
KeyPts1__ = (np.dot(deltaR, KeyPts1_.T) + deltaT.reshape(3,1)).T
pairs1__ = (np.dot(deltaR, pairs1_.T) + deltaT.reshape(3,1)).T
PlanarPts1__ = (np.dot(deltaR, PlanarPts1_.T) + deltaT.reshape(3,1)).T
Colors0=np.ones((PC0.shape[0],1), dtype=np.float32)*1.0
KeyColors0=np.ones((KeyPts0.shape[0],1), dtype=np.float32)*0.9
KeyColors0_=np.ones((KeyPts0.shape[0],1), dtype=np.float32)*0.9
PlanarColor0=np.ones((PlanarPts0.shape[0],1), dtype=np.float32)*0.7
Colors1=np.ones((PC1.shape[0],1), dtype=np.float32)*0.0
KeyColors1=np.ones((KeyPts1.shape[0],1), dtype=np.float32)*0.1
KeyColors1_=np.ones((KeyPts1.shape[0],1), dtype=np.float32)*0.1
PlanarColor1=np.ones((PlanarPts1.shape[0],1), dtype=np.float32)*0.3
Colors4FusedPC=np.r_[Colors0, Colors1]
Colors4FusedPC=Colors4FusedPC.reshape(Colors4FusedPC.shape[0],)
shift4Show = 0
fig = mlab.figure(bgcolor=(0, 0, 0), size=(1500, 900))
PtSize = 0.1
node = mlab.points3d(PC0[:,0], PC0[:,1], PC0[:,2], mode="point", figure=fig)
node.glyph.scale_mode = 'scale_by_vector'
node.mlab_source.dataset.point_data.scalars = Colors0
node = mlab.points3d(PC1_[:,0], PC1_[:,1], PC1_[:,2]+shift4Show, mode="point", figure=fig)
node.glyph.scale_mode = 'scale_by_vector'
node.mlab_source.dataset.point_data.scalars = Colors1
node = mlab.points3d(KeyPts0[:,0], KeyPts0[:,1], KeyPts0[:,2], scale_factor=PtSize, figure=fig)
node.glyph.scale_mode = 'scale_by_vector'
node.mlab_source.dataset.point_data.scalars = KeyColors0
node = mlab.points3d(KeyPts1_[:,0], KeyPts1_[:,1], KeyPts1_[:,2]+shift4Show, scale_factor=PtSize, figure=fig)
node.glyph.scale_mode = 'scale_by_vector'
node.mlab_source.dataset.point_data.scalars = KeyColors1
PtSize = 0.02
node = mlab.points3d(PlanarPts0[:,0], PlanarPts0[:,1], PlanarPts0[:,2], scale_factor=PtSize, figure=fig)
node.glyph.scale_mode = 'scale_by_vector'
node.mlab_source.dataset.point_data.scalars = PlanarColor0
node = mlab.points3d(PlanarPts1_[:,0], PlanarPts1_[:,1], PlanarPts1_[:,2]+shift4Show, scale_factor=PtSize, figure=fig)
node.glyph.scale_mode = 'scale_by_vector'
node.mlab_source.dataset.point_data.scalars = PlanarColor1
if bShowPairs == True:
mlab.quiver3d(pairs1_[:,0], pairs1_[:,1], pairs1_[:,2]+shift4Show, \
pairs0[:,0]-pairs1_[:,0], pairs0[:,1]-pairs1_[:,1], pairs0[:,2]-pairs1_[:,2]-shift4Show, \
figure=fig, line_width=0.5, scale_factor=1)
mlab.title('Initial Matching')
mlab.axes(xlabel='X', ylabel='Y', zlabel='Z')
PtSize = 0.2
fig = mlab.figure(bgcolor=(0, 0, 0), size=(1200, 800))
mlab.points3d(FusedPC__[:,0], FusedPC__[:,1], FusedPC__[:,2],
Colors4FusedPC, mode="point", figure=fig)
if bShowPairs == True:
mlab.quiver3d(pairs1__[:,0], pairs1__[:,1], pairs1__[:,2], \
pairs0[:,0]-pairs1__[:,0], pairs0[:,1]-pairs1__[:,1], pairs0[:,2]-pairs1__[:,2], \
figure=fig, line_width=0.5, scale_factor=1)
mlab.title('Fused PC: ' + str(iFrame0) + '-' + str(iFrame1),)
mlab.axes(xlabel='X', ylabel='Y', zlabel='Z')
# fig = mlab.figure(bgcolor=(0, 0, 0), size=(1200, 800))
# mlab.points3d(FusedPC___[:,0], FusedPC___[:,1], FusedPC___[:,2],
# Colors4FusedPC, mode="point", figure=fig)
# mlab.title('Fused PC - PlanarPts: ' + str(iFrame0) + '-' + str(iFrame1),)
# mlab.axes(xlabel='X', ylabel='Y', zlabel='Z')
mlab.show()
def GetErrorEulers(relRs0, relRs1):
assert relRs0.shape[0] == relRs1.shape[0]
nDatas = relRs0.shape[0]
ErrorEulers = np.zeros((nDatas,3), dtype=np.float32)
for i in range(nDatas):
errorR = dot(np.linalg.inv(relRs0[i,:,:]), relRs1[i,:,:])
eulers = RotateMat2EulerAngle_XYZ(errorR)
ErrorEulers[i,:] = eulers
return ErrorEulers
def GetErrorRTs(poses, poses_, Tr, isPlot):
# ----- then compare with the ground truth
relRs, relTs, relEulers, diffNormRelEulers, diffNormRelTs = GetLidarDiffRels(poses, Tr)
relRs_, relTs_, relEulers_, diffNormRelEulers_, diffNormRelTs_ = GetLidarDiffRels(poses_, Tr)
errorRelEulers = GetErrorEulers(relRs, relRs_)
errorRelTs = relTs_ - relTs
errorRelEulersNorm = LA.norm(errorRelEulers, axis=1)
errorRelTsNorm = LA.norm(errorRelTs, axis=1)
if isPlot == True:
plt.figure()
plt.subplot(2, 4, 1)
plt.title('relEulers')
plt.plot(relEulers,'.')
plt.subplot(2, 4, 2)
plt.title('relTs')
plt.plot(relTs,'.')
plt.subplot(2, 4, 3)
plt.title('diffNormRelEulers')
plt.plot(diffNormRelEulers,'.')
plt.subplot(2, 4, 4)
plt.title('diffNormRelTs')
plt.plot(diffNormRelTs,'.')
plt.subplot(2, 4, 5)
plt.title('relEulers_')
plt.plot(relEulers_,'.')
plt.subplot(2, 4, 6)
plt.title('relTs_')
plt.plot(relTs_,'.')
plt.subplot(2, 4, 7)
plt.title('diffNormRelEulers_')
plt.plot(diffNormRelEulers_,'.')
plt.subplot(2, 4, 8)
plt.title('diffNormRelTs_')
plt.plot(diffNormRelTs_,'.')
plt.figure()
plt.subplot(1, 2, 1)
plt.title('errorRelEulersNorm')
plt.plot(errorRelEulersNorm,'.')
plt.subplot(1, 2, 2)
plt.title('errorRelTsNorm')
plt.plot(errorRelTsNorm,'.')
# plt.figure()
# plt.subplot(1, 3, 1)
# plt.title('errorRelEulers-X')
# plt.plot(errorRelEulers[:,0],'.')
# plt.subplot(1, 3, 2)
# plt.title('errorRelEulers-Y')
# plt.plot(errorRelEulers[:,1],'.')
# plt.subplot(1, 3, 3)
# plt.title('errorRelEulers-Z')
# plt.plot(errorRelEulers[:,2],'.')
#
# plt.figure()
# plt.subplot(1, 3, 1)
# plt.title('errorRelTs-X')
# plt.plot(errorRelTs[:,0],'.')
# plt.subplot(1, 3, 2)
# plt.title('errorRelTs-Y')
# plt.plot(errorRelTs[:,1],'.')
# plt.subplot(1, 3, 3)
# plt.title('errorRelTs-Z')
# plt.plot(errorRelTs[:,2],'.')
plt.show
GroundTruthRels = []
GroundTruthRels.append(relRs)
GroundTruthRels.append(relTs)
GroundTruthRels.append(relEulers)
GroundTruthRels.append(diffNormRelEulers)
GroundTruthRels.append(diffNormRelTs)
EstimatedRels = []
EstimatedRels.append(relRs_)
EstimatedRels.append(relTs_)
EstimatedRels.append(relEulers_)
EstimatedRels.append(diffNormRelEulers_)
EstimatedRels.append(diffNormRelTs_)
return GroundTruthRels, EstimatedRels, errorRelEulers, errorRelTs