forked from renmengye/deep-tracker
-
Notifications
You must be signed in to change notification settings - Fork 0
/
kitti_new.py
173 lines (146 loc) · 5.79 KB
/
kitti_new.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
from __future__ import division
import cslab_environ
import cv2
import numpy as np
import os
import tfplus
from tracking_data_assembler import TrackingDataAssembler
from tracking_data_provider import TrackingDataProvider
tfplus.cmd_args.add('kitti:dataset_folder', 'str',
'/ais/gobi4/mren/data/kitti/tracking')
class KITTITrackingDataAssembler(TrackingDataAssembler):
def __init__(self, folder, output_fname=None, split='train'):
self.folder = folder
self.split = split
if output_fname is None:
output_fname = os.path.join(folder, '{}.h5'.format(split))
if split == 'train' or split == 'valid' or split == 'train_all':
self.left_folder = os.path.join(folder, 'training', 'image_02')
self.label_folder = os.path.join(folder, 'training', 'label_02')
elif split == 'test':
self.left_folder = os.path.join(folder, 'testing', 'image_02')
self.label_folder = None
else:
raise Exception('Unknown split "{}"'.format(split))
self.anns = {}
super(KITTITrackingDataAssembler, self).__init__(output_fname)
pass
def get_video_ids(self):
all_ids = filter(lambda x: x.startswith(
'0'), os.listdir(self.left_folder))
if self.split == 'train_all' or self.split == 'test':
return all_ids
elif self.split == 'train':
return all_ids[: 13]
elif self.split == 'valid':
return all_ids[13:]
else:
raise Exception('Unknown split "{}"'.format(split))
pass
def get_frame_ids(self, vid_id):
vid_folder = os.path.join(self.left_folder, vid_id)
return sorted(map(lambda x: x[:6], os.listdir(vid_folder)))
def get_frame_img(self, vid_id, frm_id):
fname = os.path.join(self.left_folder, vid_id, frm_id + '.png')
return cv2.imread(fname)
def _read_annotations(self, vid_id):
label_fname = os.path.join(self.label_folder, vid_id + '.txt')
# target_types = set(['Van', 'Car', 'Truck'])
target_types = set(['Car'])
obj_data = {}
idx_map = []
frame_start = None
frame_end = None
with open(label_fname) as label_f:
lines = label_f.readlines()
for ll in lines:
parts = ll.split(' ')
frame_no = int(parts[0])
ins_no = int(parts[1])
typ = parts[2]
truncated = int(parts[3])
occluded = int(parts[4])
bleft = float(parts[6])
btop = float(parts[7])
bright = float(parts[8])
bbot = float(parts[9])
if frame_start is None:
frame_start = frame_no
frame_end = frame_no
else:
frame_start = min(frame_start, frame_no)
frame_end = max(frame_start, frame_no)
raw_data = {
'frame_no': frame_no,
'ins_no': ins_no,
'typ': typ,
'truncated': truncated,
'occluded': occluded,
'bbox': (bleft, btop, bright, bbot)
}
if ins_no != -1 and typ in target_types:
if ins_no in obj_data:
obj_data[ins_no].append(raw_data)
else:
obj_data[ins_no] = [raw_data]
num_ins = len(obj_data.keys())
num_frames = frame_end - frame_start + 1
bbox = np.zeros([num_ins, num_frames, 5], dtype='float32')
for idx in obj_data.iterkeys():
new_idx = len(idx_map)
for dd in obj_data[idx]:
new_frame = dd['frame_no'] - frame_start
bbox[new_idx, new_frame, 4] = 1.0
bbox[new_idx, new_frame, 0: 4] = dd['bbox']
idx_map.append(idx)
idx_map = np.array(idx_map, dtype='uint8')
frame_map = np.arange(frame_start, frame_end + 1)
self.anns[vid_id] = bbox
pass
def get_obj_ids(self, vid_id):
if self.label_folder is None:
return None
if vid_id not in self.anns:
self._read_annotations(vid_id)
return ['{:04d}'.format(x) for x in xrange(self.anns[vid_id].shape[0])]
def get_obj_data(self, vid_id, obj_id):
if self.label_folder is None:
return None
obj_idx = int(obj_id)
results = {
'bbox': self.anns[vid_id][obj_idx, :, :4],
'presence': self.anns[vid_id][obj_idx, :, 4]
}
return results
pass
class KITTITrackingDataProvider(TrackingDataProvider):
def __init__(self, split='train', filename=None):
super(KITTITrackingDataProvider, self).__init__(
split=split, filename=filename)
self.register_option('kitti:dataset_folder')
pass
@property
def filename(self):
if self._filename is None:
return os.path.join(self.get_option('kitti:dataset_folder'),
self.split + '.h5')
else:
return self._filename
pass
tfplus.data.data_provider.register('kitti_track', KITTITrackingDataProvider)
if __name__ == '__main__':
# for split in ['train', 'valid']:
# for split in ['train', 'valid', 'test']:
for split in ['train', 'test']:
assembler = KITTITrackingDataAssembler(
'/ais/gobi4/mren/data/kitti/tracking', split=split)
# print assembler.get_frame_ids('0017')
assembler.assemble()
pass
# b = tfplus.data.create_from_main('kitti_track').get_batch_idx(np.arange(5))
# print b['x'].shape
# print b['x']
# print b['bbox_gt'].shape
# print b['bbox_gt']
# print b['s_gt'].shape
# print b['s_gt']