def load_tags(self, db_file_name):
"""
Load tags from a file.
:type db_file_name: string
:param db_file_name: the root name of the file containing the tags
:rtype: boolean
:returns: True if loading was successful
"""
if db_file_name is None:
# Set the name of database.
# Algorithm : LBP (Local Binary Pattern)
db_file_name = self._db_name
tags_file_name = self._db_name + "-LBP-Tags"
ok = False
if os.path.isfile(tags_file_name):
self._tags = utils.load_YAML_file(tags_file_name)
ok = True
print('\n### TAGS LOADED ###\n')
return ok
def load(self, db_file_name):
"""
Update the face models data structure from a file.
:type db_file_name: string
:param db_file_name: the name of the file containing
the dump of the face models data structure
:rtype: boolean
:returns: True if loading was successful
"""
if db_file_name is None:
'''
Set the name of database.
Algorithm :
LBP (Local Binary Pattern)
'''
db_file_name = self._db_name + '-' + self._algorithm
tags_file_name = db_file_name + '-Tags'
algorithm = ce.FACE_MODEL_ALGORITHM
if self._params is not None:
algorithm = self._params[ce.FACE_MODEL_ALGORITHM_KEY]
model = None
if algorithm == 'Eigenfaces':
model = cv2.createEigenFaceRecognizer()
elif algorithm == 'Fisherfaces':
model = cv2.createFisherFaceRecognizer()
elif algorithm == 'LBP':
model = cv2.createLBPHFaceRecognizer()
ok = False
if os.path.isfile(db_file_name) and (os.path.isfile(tags_file_name)):
if(not((ce.USE_TRACKING or ce.SIM_TRACKING or ce.USE_SLIDING_WINDOW)
and ce.LOAD_IND_FRAMES_RESULTS)):
model.load(db_file_name)
if not(model is None):
self.model = model
self._tags = utils.load_YAML_file(tags_file_name)
ok = True
print('\n### DB LOADED ###\n')
return ok
def load_experiment_results(file_path):
"""
Load file with results of all experiments
:type file_path: string
:param file_path: file to be loaded
:rtype: list
:returns: list with experiments
"""
data = utils.load_YAML_file(file_path)
experiments = data[ce.EXPERIMENTS_KEY]
return experiments
def load_image_annotations(file_path):
"""
Load YAML file with image .
:type file_path: string
:param file_path: path of YAML file to be loaded
:rtype: list
:returns: a list of dictionaries with the annotated images
"""
data = utils.load_YAML_file(file_path)
if data:
images = data[c.ANNOTATIONS_FRAMES_KEY]
return images
else:
print 'Unable to open', file_path
return None
if __name__ == "__main__":
import argparse
import sys
parser = argparse.ArgumentParser(
description="Execute face recognition tests")
parser.add_argument("-config", help="configuration file")
args = parser.parse_args()
params = None
if args.config:
# Load given configuration file
try:
params = load_YAML_file(args.config)
except IOError, (errno, strerror):
print("I/O error({0}): {1}".format(errno, strerror))
print("Default configuration file will be used")
except:
print("Unexpected error:", sys.exc_info()[0])
raise
print("\n ### EXECUTING SOFTWARE TEST ###\n")
test_passed = fr_test(params, False)
if test_passed:
print("\nSOFTWARE TEST PASSED\n")
print("\n ### EXECUTING EXPERIMENTS ###\n")
fr_experiments(params, False)
if __name__ == "__main__":
parser = argparse.ArgumentParser(
description="Execute clothing recognition tests")
parser.add_argument(
"-dataset_path", help="path of dataset")
parser.add_argument("-config", help="configuration_file")
args = parser.parse_args()
# Set parameters
params = None
if args.config:
# Load given configuration file
try:
params = utils.load_YAML_file(args.config)
except IOError, (errno, strerror):
print("I/O error({0}): {1}".format(errno, strerror))
print("Default configuration file will be used")
except:
print("Unexpected error:", sys.exc_info()[0])
raise
else:
print("Default configuration file will be used")
# Set paths
dataset_path = None
if args.dataset_path:
dataset_path = args.dataset_path
else:
print("Path of dataset not provided")
def person_tracking_image_experiments(params=None):
"""
Execute person tracking experiments on images
:type params: dictionary
:param params: configuration parameters
to be used for the experiment (see table)
============================================ ======================================== ==============
Key Value Default value
============================================ ======================================== ==============
person_tracking_clothes_bbox_height Height of bounding box for clothes
(in % of the face bounding box height) 1.0
person_tracking_clothes_bbox_width Width of bounding box for clothes 2.0
(in % of the face bounding box width)
person_tracking_neck_height Height of neck (in % of the 0.0
face bounding box height)
nr_of_hsv_channels_nr_in_person_tracking Number of HSV channels used 2
in person tracking (1-2)
use_mask_in_person_tracking If True, use a mask for HSV values False
min_size_height Minimum height of face detection 20
bounding box (in pixels)
min_size_width Minimum width of face detection 20
bounding box (in pixels)
annotations_path Path of directory containing the
manual annotations for the images
dataset_path Path of dataset
============================================ ======================================== ==============
"""
ann_path = ce.PERSON_TRACKING_IMAGE_ANN_PATH
dataset_path = ce.PERSON_TRACKING_IMAGE_DATASET_PATH
if params is not None:
if ce.PERSON_TRACKING_IMAGE_ANN_PATH_KEY in params:
ann_path = params[ce.PERSON_TRACKING_IMAGE_ANN_PATH_KEY]
if ce.PERSON_TRACKING_IMAGE_DATASET_PATH_KEY in params:
dataset_path = params[ce.PERSON_TRACKING_IMAGE_DATASET_PATH_KEY]
# Load file with annotations
ann_dict = utils.load_YAML_file(ann_path)
tot_discrete_score = 0
tot_cont_score = 0
tot_tracking_time = 0
matches_counter = 0
for subject_dir in os.listdir(dataset_path):
subject_path = os.path.join(dataset_path, subject_dir)
# Every image in directory is used as a reference image
for ref_im_name in os.listdir(subject_path):
print('ref_im_name', ref_im_name)
# Full path of reference image
ref_im_path = os.path.join(subject_path, ref_im_name)
# Path of image relative to dataset path
ref_rel_im_path = os.path.join(subject_dir, ref_im_name)
ref_bbox = ann_dict[ref_rel_im_path][c.BBOX_KEY]
# Other images in same directory are used as test images
for test_im_name in os.listdir(subject_path):
if test_im_name != ref_im_name:
# Full path of test image
test_im_path = os.path.join(subject_path, test_im_name)
# Path of image relative to dataset path
test_rel_im_path = os.path.join(subject_dir, ref_im_name)
test_true_bbox = ann_dict[test_rel_im_path][c.BBOX_KEY]
true_face_width = test_true_bbox[2]
true_face_height = test_true_bbox[3]
show_results = False
# Saving processing time
start_time = cv2.getTickCount()
test_found_bbox = pt.find_person_by_clothes(
test_im_path, ref_im_path, ref_bbox, params, show_results)
clocks = cv2.getTickCount() - start_time
seconds = clocks / cv2.getTickFrequency()
tot_tracking_time += seconds
if test_found_bbox:
(sim, int_area, int_area_pct) = utils.is_rect_similar(
test_true_bbox, test_found_bbox, 0)
if sim:
tot_discrete_score += 1
tot_cont_score += int_area_pct
if int_area_pct > 1.8:
print('int_area_pct', int_area_pct)
pt.find_person_by_clothes(
test_im_path, ref_im_path, ref_bbox,
params, True)
matches_counter += 1
discrete_score = float(tot_discrete_score) / matches_counter
cont_score = tot_cont_score / matches_counter
mean_tracking_time = tot_tracking_time / matches_counter
print('matches_counter', matches_counter)
print("\n ### RESULTS ###\n")
print('Mean of discrete score: ' + str(discrete_score * 100) + '%')
print('Mean of continuous score: ' + str(cont_score * 100) + '%')
print('Mean tracking time: ' + str(mean_tracking_time) + ' s\n\n')
