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
|
import numpy as np
from . import draw as draw_module
from . import label as label_module
def mask_to_bbox(masks):
bboxes = np.zeros((len(masks), 4), dtype=float)
for i, mask in enumerate(masks):
if mask.sum() == 0:
continue
where = np.argwhere(mask)
(y1, x1), (y2, x2) = where.min(0), where.max(0) + 1
bbox = y1, x1, y2, x2
bboxes[i] = bbox
return bboxes
def instances2rgb(
image,
labels,
bboxes=None,
masks=None,
captions=None,
font_size=25,
line_width=5,
boundary_width=1,
alpha=0.7,
colormap=None,
font_path=None,
):
"""Convert instances to rgb.
Parameters
----------
image: numpy.ndarray, (H, W, 3), numpy.uint8
RGB image.
labels: list of int, (N,)
Labels.
bboxes: list of numpy.ndarray, (N, 4), float
Bounding boxes.
masks: numpy.ndarray, (N, H, W), bool
Masks.
captions: list of str
Captions.
font_size: int
Font size.
line_width: int
Line width.
alpha: float
Alpha of RGB.
colormap: numpy.ndarray, (M, 3), numpy.uint8
Label id to RGB color.
Returns
-------
dst: numpy.ndarray, (H, W, 3), numpy.uint8
Visualized image.
"""
assert isinstance(image, np.ndarray)
assert image.dtype == np.uint8
assert image.ndim == 3
assert all(label_i >= 0 for label_i in labels)
n_instance = len(labels)
if masks is None:
assert bboxes is not None
masks = [None] * n_instance
if bboxes is None:
assert masks is not None
bboxes = mask_to_bbox(masks)
if captions is None:
captions = [None] * n_instance
assert len(masks) == len(bboxes) == len(captions) == n_instance
if colormap is None:
colormap = label_module.label_colormap()
dst = image
for instance_id in range(n_instance):
mask = masks[instance_id]
if mask is None or mask.sum() == 0:
continue
color_ins = colormap[1:][instance_id % len(colormap[1:])]
maskviz = mask[:, :, None] * color_ins.astype(float)
dst = dst.copy()
dst[mask] = (1 - alpha) * image[mask].astype(float) + alpha * maskviz[
mask
]
try:
import skimage.segmentation
boundary = skimage.segmentation.find_boundaries(
mask, connectivity=2
)
for _ in range(boundary_width - 1):
boundary = skimage.morphology.binary_dilation(boundary)
dst[boundary] = (200, 200, 200)
except ImportError:
pass
for instance_id in range(n_instance):
bbox = bboxes[instance_id]
label = labels[instance_id]
caption = captions[instance_id]
color_cls = colormap[label % len(colormap)]
y1, x1, y2, x2 = bbox
if (y2 - y1) * (x2 - x1) == 0:
continue
aabb1 = np.array([y1, x1], dtype=int)
aabb2 = np.array([y2, x2], dtype=int)
dst = draw_module.rectangle(
dst,
aabb1,
aabb2,
outline=color_cls,
width=line_width,
)
if caption is not None:
for loc in ["lt+", "lt"]:
y1, x1, y2, x2 = draw_module.text_in_rectangle_aabb(
src=dst,
loc=loc,
text=caption,
size=font_size,
aabb1=aabb1,
aabb2=aabb2,
font_path=font_path,
)
if (
y1 >= 0
and x1 >= 0
and y2 < dst.shape[0]
and x2 < dst.shape[1]
):
break
dst = draw_module.text_in_rectangle(
src=dst,
loc=loc,
text=caption,
size=font_size,
background=color_cls,
aabb1=aabb1,
aabb2=aabb2,
font_path=font_path,
)
return dst
|
