forked from bingykang/Fewshot_Detection
-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathvoc_label_1c.py
executable file
·85 lines (69 loc) · 2.83 KB
/
voc_label_1c.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
import xml.etree.ElementTree as ET
import pickle
import os
from os import listdir, getcwd
from os.path import join
import argparse
# parser = argparse.ArgumentParser()
# parser.add_argument('--type', type=str, choices=['1c', 'all'], required=True)
# args = parser.parse_args()
sets=[('2012', 'train'), ('2012', 'val')]
classes = ["aeroplane", "bicycle", "bird", "boat", "bottle", "bus", "car", "cat", "chair", "cow", "diningtable", "dog", "horse", "motorbike", "person", "pottedplant", "sheep", "sofa", "train", "tvmonitor"]
def convert(size, box):
dw = 1./size[0]
dh = 1./size[1]
x = (box[0] + box[1])/2.0
y = (box[2] + box[3])/2.0
w = box[1] - box[0]
h = box[3] - box[2]
x = x*dw
w = w*dw
y = y*dh
h = h*dh
return (x,y,w,h)
def convert_annotation(year, image_id, class_name):
in_file = open('VOCdevkit/VOC%s/Annotations/%s.xml'%(year, image_id))
out_file = open('VOCdevkit/VOC%s/labels_1c/%s/%s.txt'%(year, class_name, image_id), 'w')
tree=ET.parse(in_file)
root = tree.getroot()
size = root.find('size')
w = int(size.find('width').text)
h = int(size.find('height').text)
for obj in root.iter('object'):
difficult = obj.find('difficult').text
cls = obj.find('name').text
if cls != class_name or int(difficult) == 1:
continue
# cls_id = classes.index(cls)
cls_id = 0
xmlbox = obj.find('bndbox')
b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(xmlbox.find('ymin').text), float(xmlbox.find('ymax').text))
bb = convert((w,h), b)
out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')
wd = getcwd()
if not os.path.exists('voclist'):
os.mkdir('voclist')
for class_name in classes:
for year, image_set in sets:
image_ids = open('VOCdevkit/VOC%s/ImageSets/Main/%s_%s.txt'%(year, class_name, image_set)).read().strip().split()
ids, flags = image_ids[::2], image_ids[1::2]
image_ids = list(zip(ids, flags))
# File to save the image path list
list_file = open('voclist/%s_%s_%s.txt'%(year, class_name, image_set), 'w')
# File to save the image labels
label_dir = 'labels_1c/' + class_name
if not os.path.exists('VOCdevkit/VOC%s/%s/'%(year, label_dir)):
os.makedirs('VOCdevkit/VOC%s/%s/'%(year, label_dir))
# Traverse all images
for image_id, flag in image_ids:
if int(flag) == -1:
continue
list_file.write('%s/VOCdevkit/VOC%s/JPEGImages/%s.jpg\n'%(wd, year, image_id))
convert_annotation(year, image_id, class_name)
list_file.close()
files = [
'voclist/2012_{}_*.txt'.format(class_name)
]
files = ' '.join(files)
cmd = 'cat ' + files + '> voclist/{}_train.txt'.format(class_name)
os.system(cmd)