-
Notifications
You must be signed in to change notification settings - Fork 107
/
utils.py
executable file
·118 lines (94 loc) · 3.37 KB
/
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
import tensorflow as tf
import numpy as np
import logging
from config import config
from math import ceil, floor
log = logging.getLogger()
def filter_proposals(proposals, gt_bboxes):
if len(gt_bboxes) > 0:
idx = np.where(batch_iou(proposals, gt_bboxes).max(axis=1) < 0.5)[0]
proposals = proposals[idx]
return proposals, idx
def zoom_out(img, gt_bboxes, params):
X_out = np.random.uniform(1.0, params['X_out'])
h, w, _ = img.shape
if X_out - 1.0 < 5e-2:
return img, gt_bboxes
x_shift = np.random.randint(0, floor((X_out - 1) * w))
y_shift = np.random.randint(0, floor((X_out - 1) * h))
img_out = np.zeros([ceil(h * X_out), ceil(w * X_out), 3], dtype='float32')
img_out[:, :] = params['zoomout_color']
img_out[y_shift:(y_shift + h), x_shift:(x_shift + w)] = img
gt_bboxes[:, [0, 1]] += x_shift / w, y_shift / h
gt_bboxes /= X_out
return img_out, gt_bboxes
def hflip_rois(bboxes, width):
x = bboxes[:, 0]
y = bboxes[:, 1]
w = bboxes[:, 2]
h = bboxes[:, 3]
return np.stack([width - x - w, y, w, h], axis=1)
def batch_iou(proposals, gt):
bboxes = np.transpose(proposals).reshape((4, -1, 1))
bboxes_x1 = bboxes[0]
bboxes_x2 = bboxes[0]+bboxes[2]
bboxes_y1 = bboxes[1]
bboxes_y2 = bboxes[1]+bboxes[3]
gt = np.transpose(gt).reshape((4, 1, -1))
gt_x1 = gt[0]
gt_x2 = gt[0]+gt[2]
gt_y1 = gt[1]
gt_y2 = gt[1]+gt[3]
widths = np.maximum(0, np.minimum(bboxes_x2, gt_x2) -
np.maximum(bboxes_x1, gt_x1))
heights = np.maximum(0, np.minimum(bboxes_y2, gt_y2) -
np.maximum(bboxes_y1, gt_y1))
intersection = widths*heights
union = bboxes[2]*bboxes[3] + gt[2]*gt[3] - intersection
return (intersection / union)
def decode_bboxes(tcoords, anchors):
var_x, var_y, var_w, var_h = config['prior_variance']
t_x = tcoords[:, 0]*var_x
t_y = tcoords[:, 1]*var_y
t_w = tcoords[:, 2]*var_w
t_h = tcoords[:, 3]*var_h
a_w = anchors[:, 2]
a_h = anchors[:, 3]
a_x = anchors[:, 0]+a_w/2
a_y = anchors[:, 1]+a_h/2
x = t_x*a_w + a_x
y = t_y*a_h + a_y
w = tf.exp(t_w)*a_w
h = tf.exp(t_h)*a_h
x1 = tf.maximum(0., x - w/2)
y1 = tf.maximum(0., y - h/2)
x2 = tf.minimum(1., w + x1)
y2 = tf.minimum(1., h + y1)
return tf.stack([y1, x1, y2, x2], axis=1)
def encode_bboxes(proposals, gt):
prop_x = proposals[:, 0]
prop_y = proposals[:, 1]
prop_w = proposals[:, 2]
prop_h = proposals[:, 3]
gt_x = gt[:, 0]
gt_y = gt[:, 1]
gt_w = gt[:, 2]
gt_h = gt[:, 3]
diff_x = (gt_x + 0.5*gt_w - prop_x - 0.5*prop_w)/prop_w
diff_y = (gt_y + 0.5*gt_h - prop_y - 0.5*prop_h)/prop_h
if len(gt) > 0 and (np.min(gt_w/prop_w) < 1e-6 or np.min(gt_h/prop_h) < 1e-6):
print(np.min(gt_w), np.min(gt_h), np.min(gt_w/prop_w), np.max(gt_h/prop_h))
diff_w = np.log(gt_w/prop_w)
diff_h = np.log(gt_h/prop_h)
var_x, var_y, var_w, var_h = config['prior_variance']
x = np.stack([diff_x/var_x, diff_y/var_y, diff_w/var_w, diff_h/var_h],
axis=1)
return x
def print_variables(name, var_list, level=logging.DEBUG):
"""Handy tool for printing vars"""
variables = sorted([v.op.name for v in var_list])
s = "Variables to %s:\n%s" % (name, '\n'.join(variables))
if level < 0:
print(s)
else:
log.log(level, s)