def __init__(self, dataset, detector = 'yolov3', experiment_name = 'default', overwrite = False):
self.detector = detector
self.dataset = dataset
self.overwrite = overwrite
self.experiment_name = experiment_name
self.checkdir()
self.config = Config(detector)
self.DA = DA('validation', self.config)
def __init__(self, detector, det = None, name = 'SelectFrame'):
wx.Frame.__init__(self, None, title=name)
self.sizer = wx.BoxSizer(wx.VERTICAL)
self.SetSizer(self.sizer)
self.config = Config(detector)
self.DA = DA('validation', self.config)
self.fmts = ['*.jpg', '*.png', '*.bmp','*.jpeg']
self.menuBar()
self.Center()
self.query = {}
self.gallery = {}
pub.subscribe(self.queryReid, 'query')
def __init__(self,
dataset,
detector='yolov3',
experiment_name='default',
overwrite=False):
if dataset is None: return
self.detector = detector
self.dataset = dataset
self.overwrite = overwrite
self.experiment_name = experiment_name
self.checkdir()
self.config = Config(detector)
self.DA = DA('validation', self.config)
self.load_gallery()
self.load_query()
def thread_da(self, thread_name, delay):
self.logger.info("start new thread: %s" % thread_name)
da = DA('training', self.config)
while 1:
cnt = 0
while len(self.thread_task[thread_name]) == 0:
cnt += 1
if cnt > 200:
self.logger.warn("thread %s stacked" % thread_name)
cnt = 0
time.sleep(delay)
img1, box1 = self.thread_task[thread_name].pop()
image_1, gt_bbox1, gt_id, gt_loc = da(img1, box1)
self.thread_task_results[thread_name].append(
[image_1, gt_bbox1, gt_id, gt_loc])
def main(argv):
M = 'yolov3'
gpu = '0'
image_path = None
try:
opts, args = getopt.getopt(argv[1:], 'hm:g:p:',
['m=', 'gpu=', 'path='])
except getopt.GetoptError:
print(argv[0] + ' -m <M> -g <gpu>')
sys.exit(2)
for opt, arg in opts:
if opt == '-h':
print(argv[0] + ' -m <M> -g <gpu>')
elif opt in ['-m', '--M']:
M = arg
elif opt in ['-p', '--path']:
image_path = arg
elif opt in ['-g', '--gpu']:
gpu = arg
print('model: [%s] gpu: [%s]' % (M, gpu))
os.environ['CUDA_VISIBLE_DEVICES'] = gpu
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
set_session(sess)
if not os.path.exists(image_path):
raise ValueError(
'you must specify a image for testing %s' %
('' if image_path is None else 'bad image path: [%s]' %
image_path))
config = Config(M)
model_path = os.path.join(parpath, 'pretrained_weights/%s.h5' % config.M)
model = MODELS(config, model_type='detection').load_model(model_name='')
#test
da = DA('validation', config)
img = cv2.imread(image_path)
input_img, input_box, input_ids, meta = da(img, [])
cv2.imwrite('input_img.jpg', (input_img * 255).astype(np.uint8))
if config.M == 'mrcnn':
anchors = ANCHORS(config)
detection, detection_score = model.predict([
np.stack([input_img]),
np.stack([input_box]),
np.stack(meta),
np.stack([anchors.get_anchors(input_img.shape)])
])
else:
detection, detection_score = model.predict(
[np.stack([input_img]),
np.stack([input_box])])
detection = da.unmold(detection[0], meta)
image = draw(img, detection, detection_score)
print(detection_score)
print(detection)
cv2.imwrite('%s_test.jpg' % config.M, image)
print('detection results saved as: %s_test.jpg' % config.M)
def __init__(self, dataset, reid, detector='yolov3'):
self.detector = detector
self.reid = reid
self.dataset = dataset
self.config = Config(detector)
self.DA = DA('validation', self.config)
import cv2
import time
import os, sys
parpath = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
curpath = os.path.dirname(os.path.abspath(__file__))
sys.path.append(parpath)
from tools.load_weights import load_weights_by_name
from tools.config import Config
from tools.DataAugmentation import DataAugmentation as DA
sys.path.remove(parpath)
if __name__ == '__main__':
config = Config('dla_34')
model = DLASeg(config).model(model_type='detection')
load_weights_by_name(model, 'dla_34.h5')
da = DA('validation', config)
while 1:
start = time.time()
img0 = cv2.imread('test.jpg')
img, _, _, meta = da(img0, [])
dets, scores = model.predict([[img], np.zeros([1, 1, 4])])
boxes, scores = da.unmold(dets[0], meta), scores[0]
for i in range(0, boxes.shape[0]):
bbox = boxes[i][0:4]
cv2.rectangle(img0, (bbox[0], bbox[1]),
(bbox[0] + bbox[2], bbox[1] + bbox[3]), (0, 255, 0),
2)
cv2.imwrite('dets.jpg', img0)
break
def main(argv):
M = 'yolov3'
gpu = '0'
CUHK_SYSU = "/home/ronghua/Projects/data/dataset-v2/"
PRW = '/home/ronghua/Projects/data/PRW-v16.04.20/'
try:
opts, args = getopt.getopt(argv[1:], 'hm:g:p:c:',
['m=', 'gpu=', 'prw=', 'cuhk='])
except getopt.GetoptError:
print(argv[0] + ' -m <M> -g <gpu>')
sys.exit(2)
for opt, arg in opts:
if opt == '-h':
print(argv[0] + ' -m <M> -g <gpu>')
elif opt in ['-m', '--M']:
M = arg
elif opt in ['-p', '--path']:
PRW = arg
elif opt in ['-c', '--cuhk']:
CUHK_SYSU = arg
elif opt in ['-g', '--gpu']:
gpu = arg
print('model: [%s] gpu: [%s], CUHK_SYSU: [%s], PRW: [%s]' %
(M, gpu, CUHK_SYSU, PRW))
os.environ['CUDA_VISIBLE_DEVICES'] = gpu
dataset = False
if os.path.exists(CUHK_SYSU):
dataset = True
if os.path.exists(PRW):
dataset = True
if not dataset:
raise ValueError(
'you should specify at least one dataset [CUHK_SYSU or PRW]')
config = Config(M)
model_path = os.path.join(parpath, 'pretrained_weights/%s.h5' % config.M)
model = MODELS(config, model_type='detection').load_model(model_name='')
if config.M == 'mrcnn':
anchors = ANCHORS(config)
load_weights_by_name(model, model_path)
da = DA('validation', config)
if os.path.exists(CUHK_SYSU):
filepath = os.path.join(parpath, 'experiment_results',
'cuhk_%s_gallery.pkl' % config.M)
if os.path.exists(filepath):
print('cuhk_%s_gallery.pkl exists.' % config.M)
else:
print('creating cuhk-sysu gallery for %s' % config.M)
gallery = []
pool_path = os.path.join(CUHK_SYSU, 'dataset/annotation/pool.mat')
if not os.path.exists(pool_path):
raise ValueError('cannot found %s' % pool_path)
pool = loadmat(pool_path)['pool'].squeeze()
imnames = [imname[0] for imname in pool]
for imname in imnames:
img = cv2.imread(
os.path.join(CUHK_SYSU, 'dataset/Image/SSM/', imname))
input_img, input_box, input_ids, meta = da(img, [])
if config.M == 'mrcnn':
detection, scores = model.predict([
np.stack([input_img]),
np.stack([input_box]),
np.stack(meta),
np.stack([anchors.get_anchors(input_img.shape)])
])
else:
detection, scores = model.predict(
[np.stack([input_img]),
np.stack([input_box])])
detection = da.unmold(detection[0], meta)
detection[:, 2:] += detection[:, :2]
features = np.zeros([len(detection), 0])
gallery.append([imname, features, detection])
print("\r%d|%d" % (len(gallery), len(imnames)), end='')
print('')
f = open(filepath, 'wb')
pickle.dump(gallery, f)
f.close()
if os.path.exists(PRW):
filepath = os.path.join(parpath, 'experiment_results',
'prw_%s_gallery.pkl' % config.M)
if os.path.exists(filepath):
print('prw_%s_gallery.pkl exists.' % config.M)
else:
print('creating prw gallery for %s' % config.M)
gallery = []
frame_test_path = os.path.join(PRW, 'frame_test.mat')
if not os.path.exists(frame_test_path):
raise ValueError('cannot found %s' % frame_test_path)
frame_indices = loadmat(
frame_test_path)['img_index_test'].squeeze()
imnames = [imname[0] + '.jpg' for imname in frame_indices]
for imname in imnames:
img = cv2.imread(os.path.join(PRW, 'frames', imname))
input_img, input_box, input_ids, meta = da(img, [])
if config.M == 'mrcnn':
detection, scores = model.predict([
np.stack([input_img]),
np.stack([input_box]),
np.stack(meta),
np.stack([anchors.get_anchors(input_img.shape)])
])
else:
detection, scores = model.predict(
[np.stack([input_img]),
np.stack([input_box])])
detection = da.unmold(detection[0], meta)
detection[:, 2:] += detection[:, :2]
features = np.zeros([len(detection), 0])
gallery.append([imname, features, detection])
print("\r%d|%d" % (len(gallery), len(imnames)), end='')
print('')
f = open(filepath, 'wb')
pickle.dump(gallery, f)
f.close()