-
Notifications
You must be signed in to change notification settings - Fork 3
/
skeletonPlotter.py
435 lines (429 loc) · 21 KB
/
skeletonPlotter.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
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
'''
Module for display skeleton animation. There is a provision for plotting
a selected number joint position over time.
TODO: Plot bone orientation, global angles etc.
TODO: Save a realtime movie out of the animation
TODO: Save publishable image sequences out of the animation
TODO: Deploy in internet
-------------------------------------------------------------------------------
Coded by Md. Iftekhar Tanveer ([email protected])
Rochester Human-Computer Interaction (ROCHCI)
University of Rochester
-------------------------------------------------------------------------------
'''
import numpy as np
import matplotlib.pyplot as plt
import fileio as fio
import scipy.io as sio
from mpl_toolkits.mplot3d import Axes3D
import matplotlib.animation as animation
import matplotlib.cm as cm
# # TODO: change arguments to all lower case. Camelcase is difficult to remember
# class plotskeleton(object):
# #TODO: redesign
# # Initialize the visualization class
# def __init__(self,data,dataheader,boneconnection,jointid1=1,\
# jointid2=9,startendtime=[],skipframes=0):
# # startendtime denotes the time (in millisec) from where the animation
# # starts and finishes. bx (boundary index) contains the same
# # information as vector indices.
# if not startendtime:
# self.bx = (0,len(data),1+skipframes)
# else:
# assert startendtime[0]<startendtime[1]
# self.bx = (np.argmax(data[:,1]>startendtime[0]),\
# np.argmax(data[:,1]>startendtime[1]),1+skipframes)
# # Joints to display in the two plots. Make sure they are tuple or list
# # not scalar
# if isinstance(jointid1,tuple) == False and \
# isinstance(jointid1,list) == False:
# self.jointid1 = (jointid1,)
# else:
# self.jointid1 = jointid1
# if isinstance(jointid2,tuple) == False and \
# isinstance(jointid2,list) == False:
# self.jointid2 = (jointid2,)
# else:
# self.jointid2 = jointid2
# # x,y, z positions of all the 20 joints and corresponding names
# # x axis mirroring for display
# self.x = data[:,2::3]*-1
# self.xhead = dataheader[2::3]
# self.y = data[:,3::3]
# self.yhead = dataheader[3::3]
# self.z = data[:,4::3]
# self.zhead = dataheader[4::3]
# # Timestamp in millisec
# self.timeStamp = data[:,1]
# # Connection of bones
# self.boneStartIdx = boneconnection[:,0]
# self.boneEndIdx = boneconnection[:,1]
# self.numBones = len(self.boneStartIdx)
# self.elev = 0
# self.azim = -90
# # Minimum and maximum of all the data columns
# self.minVals = np.min(data[self.bx[0]:self.bx[1],:],axis=0)
# self.maxVals = np.max(data[self.bx[0]:self.bx[1],:],axis=0)
# # Setup the figure and axes
# self.fig = plt.figure('Skeleton Plotter', figsize=(15, 10), dpi=80)
# self.ax = self.fig.add_axes([0,0,0.45,1], projection='3d')
# self.axplott1 = self.fig.add_axes([0.48,0.58,0.5,0.34])
# self.axplott2 = self.fig.add_axes([0.48,0.14,0.5,0.34])
# # Draw the 2D plot
# data1,legend1,data2,legend2 = self.prepDatToPlot()
# # Draw the 2d plots of joint positions
# self.axplott1.plot(self.timeStamp[self.bx[0]:self.bx[1]]/1000.0,data1)
# self.axplott1.axis('tight')
# self.axplott1.grid('on')
# self.axplott1.legend(legend1,prop={'size':8},bbox_to_anchor=\
# (0., 1.02, 1., .102), loc=3,ncol=6, mode="expand", borderaxespad=0.)
# self.axplott2.plot(self.timeStamp[self.bx[0]:self.bx[1]]/1000,data2)
# self.axplott2.axis('tight')
# self.axplott2.grid('on')
# self.axplott2.legend(legend2,prop={'size':8},bbox_to_anchor=\
# (0., 1.02, 1., .102), loc=3,ncol=6, mode="expand", borderaxespad=0.)
# # Start Animation
# self.ani = animation.FuncAnimation(self.fig, self.update,\
# xrange(self.bx[0],self.bx[1],self.bx[2]),interval=5,\
# init_func=self.setup_plot, blit=False)
# # This function will prepare the data and legend strings for plotting
# def prepDatToPlot(self):
# dat1 = np.zeros((self.bx[1]-self.bx[0],3*len(self.jointid1)))
# leg1 = ['LegText']*(3*len(self.jointid1))
# dat2 = np.zeros((self.bx[1]-self.bx[0],3*len(self.jointid2)))
# leg2 = ['LegText']*(3*len(self.jointid2))
# for idx in xrange(len(self.jointid1)):
# dat1[:,3*idx]=self.x[self.bx[0]:self.bx[1],self.jointid1[idx]]
# dat1[:,3*idx+1]=self.y[self.bx[0]:self.bx[1],self.jointid1[idx]]
# dat1[:,3*idx+2]=self.z[self.bx[0]:self.bx[1],self.jointid1[idx]]
# leg1[3*idx]=self.xhead[self.jointid1[idx]]
# leg1[3*idx+1]=self.yhead[self.jointid1[idx]]
# leg1[3*idx+2]=self.zhead[self.jointid1[idx]]
# for idx in xrange(len(self.jointid2)):
# dat2[:,3*idx]=self.x[self.bx[0]:self.bx[1],self.jointid2[idx]]
# dat2[:,3*idx+1]=self.y[self.bx[0]:self.bx[1],self.jointid2[idx]]
# dat2[:,3*idx+2]=self.z[self.bx[0]:self.bx[1],self.jointid2[idx]]
# leg2[3*idx]=self.xhead[self.jointid2[idx]]
# leg2[3*idx+1]=self.yhead[self.jointid2[idx]]
# leg2[3*idx+2]=self.zhead[self.jointid2[idx]]
# return dat1,leg1,dat2,leg2
# # The data is being plotted with the following convension. This
# # enables us to properly rotate the plot
# # X_component od data ---> x axis of plot
# # Y_component od data ---> z axis of plot
# # Z_component od data ---> y axis of plot
# def setup_plot(self):
# self.ax.grid(True)
# self.ax.tick_params(\
# axis='both',
# which='both', # both major and minor ticks are affected
# bottom='off', # ticks along the bottom edge are off
# top='off', # ticks along the top edge are off
# left='off',
# right='off',
# labelbottom='off',
# labeltop='on',
# labelleft='on',
# labelright='off') # labels along the bottom edge are off
# self.ax.set_animated=True
# # Draw the skeleton plots
# self.lines = [self.ax.plot([self.x[self.bx[0],start],\
# self.x[self.bx[0],end]],[self.z[self.bx[0],start],\
# self.z[self.bx[0],end]],zs=[self.y[self.bx[0],start],\
# self.y[self.bx[0],end]],animated=True)[0]\
# for start,end in zip(self.boneStartIdx,self.boneEndIdx)]
# # Set up vertical lines as time marker
# yMinMax1=self.axplott1.get_ylim()
# yMinMax2=self.axplott2.get_ylim()
# self.marker1 = self.axplott1.plot([self.timeStamp[self.bx[0]]/1000.0,\
# self.timeStamp[self.bx[0]]/1000.0],yMinMax1,c='r',linewidth=2)[0]
# self.marker2 = self.axplott2.plot([self.timeStamp[self.bx[0]]/1000.0,\
# self.timeStamp[self.bx[0]]/1000.0],yMinMax2,c='r',linewidth=2)[0]
# # Set up viewing angle
# self.ax.view_init(elev=self.elev, azim=self.azim)
# # Set the x, y and z ranges of the skeleton plot
# self.ax.set_xlim3d(np.min(self.minVals[2::3]),\
# np.max(self.maxVals[2::3]))
# self.ax.set_ylim3d(np.min(self.minVals[4::3]),\
# np.max(self.maxVals[4::3]))
# self.ax.set_zlim3d(np.min(self.minVals[3::3]),\
# np.max(self.maxVals[3::3]))
# # These elements will be cleared
# todelete = self.lines[:]
# todelete.append(self.marker1)
# todelete.append(self.marker2)
# return (item for item in todelete)
# # The data is being plotted with the following convension. This
# # enables us to properly rotate the plot
# # X_component od data ---> x axis of plot
# # Y_component od data ---> z axis of plot
# # Z_component od data ---> y axis of plot
# def update(self, i):
# # Update the datapoints in the skeleton plot
# for idx,_ in enumerate(self.lines):
# self.lines[idx].set_data([self.x[i,self.boneStartIdx[idx]],\
# self.x[i,self.boneEndIdx[idx]]],\
# [self.z[i,self.boneStartIdx[idx]],\
# self.z[i,self.boneEndIdx[idx]]])
# self.lines[idx].set_3d_properties([self.y[i,\
# self.boneStartIdx[idx]],self.y[i,self.boneEndIdx[idx]]])
# plt.draw()
# # Update the marker
# self.marker1.set_data([self.timeStamp[i]/1000.0,\
# self.timeStamp[i]/1000.0],self.axplott1.get_ylim())
# self.marker2.set_data([self.timeStamp[i]/1000.0,\
# self.timeStamp[i]/1000.0],self.axplott2.get_ylim())
# # Okay, draw now
# plt.draw()
# # These elements will be cleared
# todelete = self.lines[:]
# todelete.append(self.marker1)
# todelete.append(self.marker2)
# return (item for item in todelete)
# # Start the animated plot
# def show(self):
# plt.show()
# Helper Function: name each bone in terms of joints
def skelNames(dataheader,boneconnection):
xhead = dataheader[2::3]
jointname = [str(item.split('_')[0]) for item in xhead]
# Names of the bones
bonename = [jointname[item1]+'_'+jointname[item2] \
for (item1,item2) in boneconnection]
return jointname,bonename
# Plot only a single frame on axis ax
def plotsingleframe(ax,data,dataheader,boneconnection,azim,elev,boxon=True):
# x,y,z coordinates of the joints. x axis mirroring for display
x = data[:,2::3]*-1
y = data[:,3::3]
z = data[:,4::3]
# Names of elements of skeleton
jointname,bonename = skelNames(dataheader,boneconnection)
uj = list(set(jointname))
ub = list(set(bonename))
# draw each bone in a unique color
boneColorMap = cm.nipy_spectral(np.linspace(0,1,len(boneconnection)))
for idx,(start,end) in enumerate(boneconnection):
thisbonecolor = boneColorMap[ub.index(bonename[idx])]
ax.plot([x[0,start],x[0,end]],[z[0,start],z[0,end]],\
zs=[y[0,start],y[0,end]],linewidth=2,c=thisbonecolor)
# Draw each joint in a unique color
jointColorMap = cm.hsv(np.linspace(1,0,len(x.T)))
for idx,(x_,y_,z_) in enumerate(zip(x.T,y.T,z.T)):
thisjointcolor = jointColorMap[uj.index(jointname[idx])]
if jointname[idx]=='HEAD':
ax.scatter(x_,z_,y_,s=500,c=jointColorMap[6])
else:
ax.scatter(x_,z_,y_,s=30,c=thisjointcolor)
__setaxis__(ax,azim,elev,boxon)
def __setaxis__(ax,azim,elev,boxon):
# Setting up view related parameters
ax.grid(False)
ax.tick_params(\
axis='both',
which='both', # both major and minor ticks are affected
bottom='off', # ticks along the bottom edge are off
top='off', # ticks along the top edge are off
left='off',
right='off',
labelbottom='off',
labeltop='off',
labelleft='on',
labelright='off') # labels along the bottom edge are off
ax.view_init(elev=elev, azim=azim)
ax.set_aspect(5)
plt.xlabel('azimuth = '+str(abs(azim)))
if not boxon:
ax.set_axis_off()
# Plot Multiple frames on axis ax
def plotmultiframe(ax,multidata,spacing,dataheader,boneconnection,azim,elev,boxon=True):
# x,y,z coordinates of the joints. x axis mirroring for display
for id,data in enumerate(multidata):
data=data[None]
x = data[:,2::3]*-1 + id*spacing
y = data[:,3::3]
z = data[:,4::3]
# Names of elements of skeleton
jointname,bonename = skelNames(dataheader,boneconnection)
uj = list(set(jointname))
ub = list(set(bonename))
# draw each bone in a unique color
boneColorMap = cm.nipy_spectral(np.linspace(0,1,len(boneconnection)))
for idx,(start,end) in enumerate(boneconnection):
thisbonecolor = boneColorMap[ub.index(bonename[idx])]
ax.plot([x[0,start],x[0,end]],[z[0,start],z[0,end]],\
zs=[y[0,start],y[0,end]],linewidth=2,c=thisbonecolor,alpha=(id+1.)/(len(multidata)+1))
# Draw each joint in a unique color
jointColorMap = cm.hsv(np.linspace(1,0,len(x.T)+1))
for idx,(x_,y_,z_) in enumerate(zip(x.T,y.T,z.T)):
thisjointcolor = jointColorMap[uj.index(jointname[idx])]
if jointname[idx]=='HEAD':
ax.scatter(x_,z_,y_,s=500,c=jointColorMap[6],alpha=(id+1.)/(len(multidata)+1))
else:
ax.scatter(x_,z_,y_,s=30,c=thisjointcolor,alpha=(id+1.)/(len(multidata)+1))
# Setting up view related parameters
__setaxis__(ax,azim,elev,boxon)
# Plot Multiple frames on axis ax with Selective Bone Highlight feature
def plotmultiframe_SBH(ax,multidata,spacing,dataheader,boneconnection,\
azim,elev,highlightedBones,boxon=True,zspacing=False):
# x,y,z coordinates of the joints. x axis mirroring for display
for id,data in enumerate(multidata):
data=data[None]
x = data[:,2::3]*-1 + id*spacing
y = data[:,3::3]
if zspacing:
z = data[:,4::3] + id*spacing*0.2
else:
z = data[:,4::3]
# Names of elements of skeleton
jointname,bonename = skelNames(dataheader,boneconnection)
uj = list(set(jointname))
ub = list(set(bonename))
# draw each bone in a unique color
boneColorMap = cm.bone(np.linspace(0,1,len(boneconnection)))
for idx,(start,end) in enumerate(boneconnection):
thisbonecolor = boneColorMap[ub.index(bonename[idx])]
if bonename[idx] in highlightedBones:
ax.plot([x[0,start],x[0,end]],[z[0,start],z[0,end]],\
zs=[y[0,start],y[0,end]],linewidth=2,c=thisbonecolor,alpha=1.0)
else:
ax.plot([x[0,start],x[0,end]],[z[0,start],z[0,end]],\
zs=[y[0,start],y[0,end]],linewidth=2,c=thisbonecolor,alpha=0.2)
# Draw each joint in a unique color
jointColorMap = cm.hsv(np.linspace(1,0,len(x.T)+1))
for idx,(x_,y_,z_) in enumerate(zip(x.T,y.T,z.T)):
thisjointcolor = jointColorMap[uj.index(jointname[idx])]
if jointname[idx]=='HEAD':
ax.scatter(x_,z_,y_,s=500,c=jointColorMap[6],alpha=0.2)
else:
ax.scatter(x_,z_,y_,s=30,c=thisjointcolor,alpha=0.2)
# Setting up view related parameters
__setaxis__(ax,azim,elev,boxon)
# Plot only a single frame on axis ax
def plotsingleframe_multiangle(data,dataheader,boneconnection,azims,elev,\
figurename='multiangle plot',boxon=True):
N = len(azims)+2
fig = plt.figure(figurename)
for i,angl in enumerate(azims):
ax = fig.add_axes([(i+1)*1./N,0,1./N,1], projection='3d')
plotsingleframe(ax,data,dataheader,boneconnection,angl,elev,boxon)
plt.show()
# This function is used when there is no frame or timestamp associated with
# the joint movement data. It constructs the animation data with the assumption
# of a specific framerate and displays it. Alternatively it is possible to
# return the data instead of displaying it
# Framerate is in frames per second
def plotJointsOnly(X,framerate=30,noShow=False):
N,D = np.shape(X)
framestep = 1000/framerate # milliseconds per frame
# Prepare data
dataheader=fio.splitcsvfile(*fio.readdatafile('Data/10.1.csv'))[1]
boneconnection = fio.readskeletaltree('Data/KinectSkeleton.tree')[1]
data = np.zeros((N,D+2))
data[:,2:] = X
data[:,0] = range(N)
data[:,1] = data[:,0]*framestep
if not noShow:
# Plot skeleton
gui = plotskeleton(data,dataheader,boneconnection)
gui.show()
else:
return data
############################## Test Modules ####################################
# Reads a file and plots it
def unittest1():
data,dataheader=fio.splitcsvfile(*fio.readdatafile('Data/20.1.csv'))[0:2]
boneconnection = fio.readskeletaltree('Data/KinectSkeleton.tree')[1]
a = plotskeleton(data,dataheader,boneconnection,jointid2=(9,10),\
skipframes=5,startendtime=[5000,25000])
a.show()
# Visualizes the results
def unittest2(filename):
framerate = 6
# Read result file
allData = sio.loadmat(filename)
# Print nonzero component indices
sumAlpha = np.sum(allData['alpha_recon'],axis=0)
validIdx = np.nonzero(sumAlpha)
while True:
print 'Available nonzero components are:'
for ind in validIdx:
print ind,
print
# print 'sum of alpha for these components are:'
# for sumAlpha in sumAlpha[validIdx]:
# print '{:0.2}'.format(sumAlpha),
# print
component = input('which component do you want to see?')
# Visualize
plt.clf()
plt.plot(allData['alpha_recon'][:,component])
plt.xlabel('frame')
plt.ylabel('alpha')
# Show the animation
X = allData['psi_recon'][:,:,component]
plotJointsOnly(X,framerate)
plt.clf()
# Load mean skeleton and draw publishable plot from three different angles
def unittest3(meanfile,boxon1=True):
data = sio.loadmat(meanfile)
boneconnection = fio.readskeletaltree('Data/KinectSkeleton.tree')[1]
plotsingleframe_multiangle(data['avgSkel'],data['header'],boneconnection,\
azims=[-45,-90,-135],elev=5,figurename='Average Pose',boxon=boxon1)
# Load a component from the result and draw publishable plot illustrating
# the action
def unittest4(filename, actionidx):
allData = sio.loadmat(filename)
boneconnection = fio.readskeletaltree('Data/KinectSkeleton.tree')[1]
header = [str(head.strip()) for head in allData['Header']]
data = fio.calcinvarient(plotJointsOnly(allData['psi_recon'][:,:,actionidx],\
framerate=30/allData['decimateratio'],noShow=True),header)
N = len(data)
fig = plt.figure('Action')
fig.set_facecolor('white')
for i,adata in enumerate(data):
ax = fig.add_axes([i*1./N,0,1./N,1], projection='3d')
plotsingleframe(ax,adata[None],allData['Header'],boneconnection,-90,5,False)
plt.show()
# Load a component from the result and draw publishable plot illustrating
# the action. In another format
def unittest5(filename, actionidx):
allData = sio.loadmat(filename)
boneconnection = fio.readskeletaltree('Data/KinectSkeleton.tree')[1]
header = [str(head.strip()) for head in allData['Header']]
data = fio.calcinvarient(plotJointsOnly(allData['psi_recon'][:,:,actionidx],\
framerate=30/allData['decimateratio'],noShow=True),header)
fig = plt.figure('Action')
ax = fig.add_subplot(111, projection='3d')
plotmultiframe(ax,data,4,allData['Header'],boneconnection,-100,5,False)
plt.show()
# Load a component from the result and draw publishable plot illustrating
# the action. Utilize bone highlight feature
def unittest6(filename, actionidx,azim,elev,highlightedBones=[\
'SHOULDER_ELBOW','ELBOW_WRIST','WRIST_HAND'],space=4,boxon=False,zspacing=False):
allData = sio.loadmat(filename)
boneconnection = fio.readskeletaltree('Data/KinectSkeleton.tree')[1]
header = [str(head.strip()) for head in allData['Header']]
data = fio.calcinvarient(plotJointsOnly(allData['psi_recon'][:,:,actionidx],\
framerate=30/allData['decimateratio'],noShow=True),header)
fig = plt.figure('Action')
ax = fig.add_subplot(111, projection='3d')
plotmultiframe_SBH(ax,data,space,allData['Header'],boneconnection,azim,\
elev,highlightedBones,boxon,zspacing)
plt.show()
if __name__ == '__main__':
#unittest1()
#unittest2('Results/result_3__M=64_D=16_beta=0.055_04_58_38.mat')
unittest3('Data/meanSkel.mat',True)
#unittest4('Results/top8_all/result_M=8_D=12_beta=4.5e-07_ALL_20_42_35.mat',1)
#unittest5('Results/top8_all/result_M=8_D=12_beta=4.5e-07_ALL_20_42_35.mat',1)
#unittest6('Results/top8_all/result_M=8_D=12_beta=4.5e-07_ALL_20_42_35.mat',1,-97,5)
#unittest6('Results/top8_all/result_M=8_D=12_beta=4.5e-07_ALL_20_42_35.mat',3,-90,25)
#unittest6('Results/top8_all/result_M=8_D=12_beta=4.5e-07_ALL_20_42_35.mat',\
#5,-85,-160,highlightedBones=['HIP_KNEE','KNEE_ANKLE','ANKLE_FOOT'],space=10,zspacing = False)
#unittest6('Results/top8_all/result_M=8_D=12_beta=4.5e-07_ALL_20_42_35.mat',\
#7,94,178,space=8,highlightedBones=['HIP_KNEE','KNEE_ANKLE','ANKLE_FOOT','SHOULDER_ELBOW','ELBOW_WRIST'],boxon=False)
#unittest6('Results/top8_all/result_M=8_D=12_beta=4.5e-07_ALL_20_42_35.mat',\
#8,-85,-158,space=4)
#unittest6('Results/top8_all/result_M=8_D=12_beta=4.5e-07_ALL_20_42_35.mat',\
#11,87,2,space=3,highlightedBones=['HIP_KNEE','KNEE_ANKLE','ANKLE_FOOT'],boxon=False)