-
Notifications
You must be signed in to change notification settings - Fork 0
/
utils_a2d.py
211 lines (177 loc) · 7 KB
/
utils_a2d.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
#!/usr/bin/env python
# Sadjad Esfeden, sadjad@ece.neu.edu
# 05/20/2016
import numpy as np
import struct
def a2d_classes():
classes = {#'adult-climbing' : 1, 'adult-crawling' : 2, 'adult-eating': 3, 'adult-jumping': 4, 'adult-rolling': 5, 'adult-running': 6, 'adult-walking': 7, 'adult-none': 8,
#'baby-climbing' : 9, 'baby-crawling' : 10, 'baby-rolling': 11, 'baby-walking': 12, 'baby-none': 13,
#'ball-flying': 14, 'ball-jumping': 15, 'ball-rolling': 16, 'ball-none': 17,
#'bird-climbing': 18, 'bird-eating': 19, 'bird-flying': 20, 'bird-jumping': 21, 'bird-rolling': 22, 'bird-walking': 23,'bird-none': 24,
#'car-flying': 25, 'car-jumping': 26, 'car-rolling': 27, 'car-running': 28, 'car-none': 29,
#'cat-climbing': 30, 'cat-eating': 31, 'cat-jumping': 32, 'cat-rolling': 33, 'cat-running': 34, 'cat-walking': 35, 'cat-none': 36,
#'dog-crawling': 37, 'dog-eating': 38, 'dog-jumping': 39, 'dog-rolling': 40, 'dog-running': 41, 'dog-walking': 42, 'dog-none': 43}
'adult' : 1, 'baby' : 2, 'ball' : 3, 'bird' : 4,
'car' : 5, 'cat' : 6, 'dog' : 7}
return classes
def a2d_palette():
palette = {( 0, 0, 0) : 0 ,
(52, 1, 1) : 1 ,
(103, 1, 1): 1,#2 , #1
(154, 1, 1): 1,#3,
(255, 1, 1): 1,#4,
(255, 51, 51): 1,#5,
(255, 103, 103): 1,#6,
(255, 154, 154): 1,#7,
(255, 205, 205): 1,#8,
(52, 46, 1) : 2,#9 ,
(103, 92, 1) : 2,#10 ,
(255, 235, 51) : 2,#11 ,
(255, 245, 154) : 2,#12 ,
(255, 250, 205) : 2,#13 ,
(41, 205, 1) : 3,#14 , #ball flying
(52, 255, 1) : 3,#15,
(92, 255, 51) : 3,#16,
(215, 255, 205) : 3,#17,
(1, 52, 36) : 4,#18 , #bird-climbing
(1, 154, 108) : 4,#19 ,
(1, 205, 143) : 4,#20 ,
(1, 255, 179) : 4,#21 ,
(51, 255, 194) : 4,#22 ,
(154, 255, 225) : 4,#23 ,
(205, 255, 240) : 4,#24 ,
(1, 82, 205) : 5,#25 ,
(1, 103, 255) : 5,#26 ,
(51, 133, 255) : 5,#27 ,
(103, 164, 255) : 5,#28 ,
(205, 225, 255) : 5,#29 ,
(26, 1, 52) : 6,#30 ,
(77, 1, 154) : 6,#31 ,
(128, 1, 255) : 6,#32 ,
(154, 51, 255) : 6,#33 ,
(179, 103, 255) : 6,#34 ,
(205, 154, 255) : 6,#35 ,
(230, 205, 255) : 6,#36 ,
(103, 1, 62) : 7,#37 ,
(154, 1, 92) : 7,#38 ,
(255, 1, 153) : 7,#39 ,
(255, 51, 174) : 7,#40 ,
(255, 103, 194) : 7,#41 ,
(255, 154, 215) : 7,#42 ,
(255, 205, 235) : 7}#,#43 }
return palette
'''{( 0, 0, 0) : 0 ,
(128, 0, 0) : 1 ,
( 0, 128, 0) : 2 ,
(128, 128, 0) : 3 ,
( 0, 0, 128) : 4 ,
(128, 0, 128) : 5 ,
( 0, 128, 128) : 6 ,
(128, 128, 128) : 7 ,
( 64, 0, 0) : 8 ,
(192, 0, 0) : 9 ,
( 64, 128, 0) : 10,
(192, 128, 0) : 11,
( 64, 0, 128) : 12,
(192, 0, 128) : 13,
( 64, 128, 128) : 14,
(192, 128, 128) : 15,
( 0, 64, 0) : 16,
(128, 64, 0) : 17,
( 0, 192, 0) : 18,
(128, 192, 0) : 19,
( 0, 64, 128) : 20 }'''
def a2d_palette_invert():
palette_list = a2d_palette().keys()
palette = ()
for color in palette_list:
palette += color
return palette
def a2d_mean_values():
return np.array([103.939, 116.779, 123.68], dtype=np.float32)
def strstr(str1, str2):
if str1.find(str2) != -1:
return True
else:
return False
# Mat to png conversion for http://www.cs.berkeley.edu/~bharath2/codes/SBD/download.html
# 'GTcls' key is for class segmentation
# 'GTinst' key is for instance segmentation
def mat2png_hariharan(mat_file, key='GTcls'):
mat = scipy.io.loadmat(mat_file, mat_dtype=True, squeeze_me=True, struct_as_record=False)
return mat[key].Segmentation
def convert_segmentation_mat2numpy(mat_file):
np_segm = load_mat(mat_file)
return np.rot90(np.fliplr(np.argmax(np_segm, axis=2)))
def load_mat(mat_file, key='data'):
mat = scipy.io.loadmat(mat_file, mat_dtype=True, squeeze_me=True, struct_as_record=False)
return mat[key]
# Python version of script in code/densecrf/my_script/LoadBinFile.m
def load_binary_segmentation(bin_file, dtype='int16'):
with open(bin_file, 'rb') as bf:
rows = struct.unpack('i', bf.read(4))[0]
cols = struct.unpack('i', bf.read(4))[0]
channels = struct.unpack('i', bf.read(4))[0]
num_values = rows * cols # expect only one channel in segmentation output
out = np.zeros(num_values, dtype=np.uint8) # expect only values between 0 and 255
for i in range(num_values):
out[i] = np.uint8(struct.unpack('h', bf.read(2))[0])
return np.rot90(np.fliplr(out.reshape((cols, rows))))
def convert_from_color_segmentation(arr_3d):
arr_2d = np.zeros((arr_3d.shape[0], arr_3d.shape[1]), dtype=np.uint8)
palette = a2d_palette()
# slow!
for i in range(0, arr_3d.shape[0]):
for j in range(0, arr_3d.shape[1]):
key = (arr_3d[i,j,0], arr_3d[i,j,1], arr_3d[i,j,2])
arr_2d[i, j] = palette.get(key, 0) # default value if key was not found is 0
return arr_2d
def create_lut(class_ids, max_id=256):
# Index 0 is the first index used in caffe for denoting labels.
# Therefore, index 0 is considered as default.
lut = np.zeros(max_id, dtype=np.uint8)
new_index = 1
for i in class_ids:
lut[i] = new_index
new_index += 1
return lut
def get_id_classes(classes):
all_classes = a2d_classes()
id_classes = [all_classes[c] for c in classes]
return id_classes
######################
'''
def palette_demo():
palette_list = a2d_palette().keys()
palette = ()
for color in palette_list:
palette += color
return palette
def convert_from_color_segmentation(arr_3d):
arr_2d = np.zeros((arr_3d.shape[0], arr_3d.shape[1]), dtype=np.uint8)
palette = a2d_palette()
# slow!
for i in range(0, arr_3d.shape[0]):
for j in range(0, arr_3d.shape[1]):
key = (arr_3d[i,j,0], arr_3d[i,j,1], arr_3d[i,j,2])
arr_2d[i, j] = palette.get(key, 0) # default value if key was not found is 0
return arr_2d
def get_id_classes(classes):
all_classes = a2d_classes()
id_classes = [all_classes[c] for c in classes]
return id_classes
def strstr(str1, str2):
if str1.find(str2) != -1:
return True
else:
return False
def create_lut(class_ids, max_id=256):
# Index 0 is the first index used in caffe for denoting labels.
# Therefore, index 0 is considered as default.
lut = np.zeros(max_id, dtype=np.uint8)
new_index = 1
for i in class_ids:
lut[i] = new_index
new_index += 1
return lut
'''