def get_list_file(file_dir):
#train_feature_list = file_io.get_listfile(file_dir + "Resized_images/","resnet_hypercolumn")
train_feature_list = file_io.get_listfile(file_dir + "Resized_images/",
"jpg")
train_label = file_io.get_listfile(file_dir + "Resized_GTdensity/", "npy")
train_list = list()
for tf in train_feature_list:
#tf_label = tf.replace(".resnet_hypercolumn","dots.png.npy").replace("images","GTdensity")
tf_label = tf.replace(".jpg",
"dots.png.npy").replace("images", "GTdensity")
assert (tf_label in train_label)
train_list.append(tf + " " + tf_label)
return train_list
def sample(avid_dir_list):
avi_dir_list = file_io.get_dir_list(avid_dir_list)
print(avi_dir_list)
for avi_dir in avi_dir_list:
avi_file_list = file_io.get_listfile(avi_dir, ".avi")
avi_file_list.sort()
for avi in avi_file_list:
image_dir = avi.replace(".avi", "")
command = "ffmpeg -i " + avi + " " + image_dir + "/%06d.jpg"
os.system(command)
def main(_):
maybe_download_and_extract()
video_dir = "/home/mscvadmin/traffic_video_analysis/data/Cam691/"
video_list = file_io.get_listfile(video_dir, ".avi")
for video in video_list:
image_dir = video.replace(".avi", "/")
name_list = os.listdir(image_dir)
sess = tf.Session()
feature_tensor = init_tensor(sess)
for n in name_list:
if n.endswith("_resize.jpg"):
image = (FLAGS.image_file if FLAGS.image_file else
os.path.join(FLAGS.model_dir, image_dir + n))
feature_tensor_v = run_sess(sess, feature_tensor, image)
feature_name = image_dir + n.replace(".jpg", ".mixed10")
save_feature(feature_tensor_v, feature_name)
mask = np.array(img)
return mask
if __name__ == "__main__":
mask_dir_list = file_io.get_dir_list("data/Cam253")
mask_list = []
for mask_dir in mask_dir_list:
if (mask_dir.endswith(".msk")):
mask_list.append(mask_dir)
print(mask_list[0])
for mask in mask_list:
image_dir_name = mask.replace(".msk", "")
image_list = file_io.get_listfile(image_dir_name, "jpg")
if (mask == "data/data_new/Training_Data/Cam181/01.msk"):
mask_bin = cv2.imread(mask_dir + "/01_mask.jpg")
mask_bin = mask_bin[:, :, 1]
mask_bin /= 255
else:
try:
mask_pts = load_mask(mask)
mask_bin = gen_mask_image(mask_pts)
except:
mask_pts = load_mask(mask)
print(mask)
print("mask is wrong")
exit(1)
continue
import file_io
if __name__ == "__main__":
file_list_dir = "../file_list/"
data_ext = "_resize.jpg"
label_ext = "_resize.desmap"
file_dir = "../data/"
cam_dir_list = file_io.get_dir_list(file_dir)
data_list = list()
for cam_dir in cam_dir_list:
video_list = file_io.get_listfile(cam_dir, ".avi")
for file_name in video_list:
file_dir_name = file_name.replace(".avi", "/")
data_list += file_io.get_listfile(file_dir_name, data_ext)
file_io.save_file(data_list, file_list_dir + 'image_name_list.txt', False)
import scipy.io as sio
import numpy as np
import file_io
import cv2
mat_dir = "../data/toJeff/Train/Resized_GTdensity/"
mat_list = file_io.get_listfile(mat_dir, "mat")
for f in mat_list:
mat = sio.loadmat(f)
np_array = mat['gtDensities']
np_array = np_array.astype(np.float32)
np_array = cv2.resize(np_array, (227,227))
np_name = f.replace(".mat",".npy")
np_array.tofile(np_name)