def save(self):
save_path = QFileDialog.getSaveFileName(self, "Save as", self.DEFAULT_TFRECORDS_SAVE_DIR, "TFRecords (*.tfrecords)")
save_path = os.path.splitext(save_path[0])[0]
if not save_path:
return
config_file = save_path + ".pkl"
pbtxt_text = self.pbtxt_text.text()
if pbtxt_text and os.path.isfile(pbtxt_text):
self.pbtxt_path = pbtxt_text
else:
self.pbtxt_path = save_path + '.pbtxt'
label_list = self.get_label_list()
with open(self.pbtxt_path, 'w') as f:
for index, label in enumerate(label_list):
f.writelines("item {\n id: %s\n name: \'%s\'\n}\n\n\n" % (index + 1, label))
itemsTextList = [str(self.list.item(i).text().split(self.DELIMITER)[0]) for i in range(self.list.count())]
valid_items = 0
dataset_configs = []
for item in itemsTextList:
path = str(item)
if os.path.exists(path):
annotation_path = self.get_roi_annotation_dir(path)
roi = DatasetDetailsDialog.getROIfromAnnotationsPath(annotation_path)
if roi is None:
roi = [-1, -1, -1, -1]
else:
roi = [roi['xmin'], roi['ymin'], roi['xmax'], roi['ymax']]
frames_list_path = os.path.join(annotation_path, '.frames_lists.pkl')
if os.path.exists(frames_list_path):
frames_list = set()
frames_list_rules = _pickle.load(open(frames_list_path, 'rb'))
for start_frame, end_frame, interval in frames_list_rules:
frames_list.update(list(range(start_frame, end_frame + 1, interval)))
frames_list = sorted(list(frames_list))
else:
frames_reader = get_frames_reader(path)
frames_list = list(range(0, frames_reader.num_frames, 1))
del frames_reader
bbox_source_filter = []
if self.use_ground_truth_bboxes.isChecked():
bbox_source_filter += ["ground_truth"]
if self.use_detection_bboxes.isChecked():
bbox_source_filter += ["object_detector"]
if self.use_tracking_bboxes.isChecked():
bbox_source_filter += ["tracker"]
dataset_configs.append(dict(
frames_source=path,
annotation_source=annotation_path,
annotation_type='pascal_voc',
label_map_file=self.pbtxt_path,
mot_class_name='object',
frames_list=frames_list,
bbox_source_filter=bbox_source_filter,
roi=roi,
))
valid_items += 1
_pickle.dump(dataset_configs, open(config_file, 'wb'))
if valid_items:
self.save_path = save_path
self.call_program(config_file, save_path + '.tfrecords')
def add_files(self):
path = os.path.expanduser('~')
opened_files = QFileDialog.getOpenFileNames(self, 'Add files', path, 'Video files (*)')
for file in opened_files[0]:
frames_reader = get_frames_reader(file)
if frames_reader.num_frames > 0:
self.addItem(file)
del frames_reader
def add_dirs(self):
path = os.path.expanduser('~')
opened_directory = QFileDialog.getExistingDirectory(self,
'Add Directory', path, QFileDialog.ShowDirsOnly
| QFileDialog.DontResolveSymlinks)
frames_reader = get_frames_reader(opened_directory)
if frames_reader.num_frames > 0:
self.addItem(opened_directory)
del frames_reader
def edit_dataset(self):
selected_dataset = self.list.currentItem().text().split(self.DELIMITER)[0]
self.currently_editing_dataset = selected_dataset
self.currently_editing_index = self.list.currentRow()
frames_reader = get_frames_reader(selected_dataset)
num_frames = frames_reader.num_frames
del frames_reader
dataset_frame_selection_dialog = \
DatasetDetailsDialog(parent=self, parent_window=self,
num_frames=num_frames, dataset_path=selected_dataset)
dataset_frame_selection_dialog.show()
def patchTracking(self, request=None, img_path=''):
request_path = request["path"]
request_roi = request["roi"]
id_number = request['id_number']
# init_frame_id = request["frame_number"]
init_frame_id = 0
init_bbox = request["bbox"]
label = request['label']
request_port = request["port"]
if request_path != self.current_path:
self.frames_reader = get_frames_reader(
request_path, save_as_bin=self.params.save_as_bin)
if request_roi is not None:
self.frames_reader.setROI(request_roi)
self.current_path = request_path
n_frames = self.frames_reader.num_frames
if self.params.end_frame_id >= init_frame_id:
end_frame_id = self.params.end_frame_id
else:
end_frame_id = n_frames - 1
for frame_id in range(init_frame_id, end_frame_id + 1):
try:
curr_frame = self.frames_reader.get_frame(frame_id)
except IOError as e:
print('{}'.format(e))
break
if request_port is not None:
self.send(curr_frame, init_bbox, label, request_path, frame_id,
id_number, request_port)
# self.single_object_tracking_results.append(tracking_result)
sys.stdout.write('Closing tracker...\n')
sys.stdout.flush()
def saveMasks(seq_path,
xml_path,
out_mask_size,
out_border,
fixed_ar,
save_raw_mask,
show_img,
out_root_path='',
save_test=1,
save_train=1,
frames_reader=None,
masks_per_seq=0,
enable_out_suffix=1,
train_fnames=None,
test_fnames=None,
map_to_bbox=0,
out_img_dir='',
enable_xml_annotations=0,
allow_skipping_images=0):
global _pause, _exit
if not xml_path or not os.path.isdir(xml_path):
raise IOError(
'Folder containing the loaded boxes does not exist: {}'.format(
xml_path))
files = glob.glob(os.path.join(xml_path, '*.xml'))
n_files = len(files)
if n_files == 0:
raise IOError('No loaded boxes found')
if frames_reader is None:
frames_reader = get_frames_reader(seq_path, save_as_bin=False)
min_dim = max_dim = 0
out_w, out_h = out_mask_size
print('out_mask_size: {}'.format(out_mask_size))
if out_w == -1 and out_h == -1:
out_w = out_h = 0
if out_w == -1:
max_dim = out_h
elif out_h == -1:
min_dim = out_w
if fixed_ar:
print('Using fixed aspect ratio: {}'.format(fixed_ar))
print('out_border: {}'.format(out_border))
def getint(fn):
basename = os.path.basename(fn)
num = re.sub("\D", "", basename)
try:
return int(num)
except:
return 0
if len(files) > 0:
files = sorted(files, key=getint)
print('Loading annotations from {:d} files'.format(n_files))
file_id = 0
n_boxes = 0
seq_root_dir = os.path.dirname(seq_path)
seq_name = os.path.basename(seq_path)
if not out_root_path:
out_root_path = os.path.join(seq_root_dir, 'masks')
if not enable_out_suffix:
out_seq_name = seq_name
else:
if map_to_bbox:
out_seq_name = '{}_mapped'.format(seq_name)
else:
out_seq_name = '{}_{}x{}'.format(seq_name, out_w, out_h)
if fixed_ar:
out_seq_name = '{}_ar_{}'.format(out_seq_name, fixed_ar)
else:
out_seq_name = '{}_{}'.format(out_seq_name, out_border)
out_seq_name = out_seq_name.replace('.', 'p')
train_root_path = os.path.join(out_root_path, out_seq_name)
if not save_test and not save_train:
raise AssertionError('Either save_test or save_train must be on')
# print('Saving output sequences to {}'.format(out_root_path))
if save_train:
out_img_root_path = train_root_path
if out_img_dir:
out_img_root_path = os.path.join(out_img_root_path, out_img_dir)
out_mask_root_path = os.path.join(train_root_path, 'labels')
print('Saving training mask sequence to {}'.format(train_root_path))
if not os.path.isdir(out_img_root_path):
os.makedirs(out_img_root_path)
if not os.path.isdir(out_mask_root_path):
os.makedirs(out_mask_root_path)
if enable_xml_annotations:
out_xml_path = os.path.join(out_img_root_path, 'annotations')
print('Saving xml_annotations to {}'.format(out_xml_path))
if not os.path.isdir(out_xml_path):
os.makedirs(out_xml_path)
if save_test:
out_test_seq_name = out_seq_name + '_test'
test_img_root_path = os.path.join(out_root_path, out_test_seq_name)
print('Saving unlabeled testing mask sequence to {}'.format(
test_img_root_path))
if not os.path.isdir(test_img_root_path):
os.makedirs(test_img_root_path)
win_name = 'patch and mask'
disable_resizing = 0
scale_x = scale_y = 1.0
if out_w == 0 and out_h == 0:
print('Resizing disabled')
disable_resizing = 1
csv_raw = []
test_csv_raw = []
n_files = len(files)
if save_raw_mask:
print('Saving raw labels')
mask_pix_val = (1, 1, 1)
else:
mask_pix_val = (255, 255, 255)
n_masks = 0
_train_fnames = []
_test_fnames = []
_exit_seq = 0
disp_img = None
for file_id, file in enumerate(files):
xml_reader = PascalVocReader(file)
filename = os.path.basename(xml_reader.filename)
filename_no_ext = os.path.splitext(filename)[0]
# file_id = int(re.sub("\D", "", filename))
# print('filename: {}'.format(filename))
# print('file_id: {}'.format(file_id))
img = frames_reader.get_frame_by_name(filename, convert_to_rgb=0)
if img is None:
print('image {} could not be read'.format(filename))
continue
img_h, img_w = img.shape[:2]
mask_img = None
shapes = xml_reader.getShapes()
n_shapes = len(shapes)
if n_shapes > 1:
print('{} boxes found for {} in {}'.format(n_shapes, filename,
file))
obj_id = 0
img_written = 0
for shape in shapes:
label, points, _, _, difficult, bbox_source, id_number, score, mask, mask_img = shape
if not mask:
if not save_test:
continue
xmin, ymin = points[0]
xmax, ymax = points[2]
img_root_path = test_img_root_path
else:
if not save_train:
continue
mask_pts_list = Shape.getContourPts(mask, verbose=0)
mask_pts = np.asarray(mask_pts_list)
xmin, ymin = np.min(mask_pts, axis=0).astype(np.int32)
xmax, ymax = np.max(mask_pts, axis=0).astype(np.int32)
img_root_path = out_img_root_path
if fixed_ar:
w, h = xmax - xmin, ymax - ymin
src_ar = float(w) / float(h)
if fixed_ar > src_ar:
border_x = int((h * fixed_ar - w) / 2.0)
border_y = 0
else:
border_y = int((w / fixed_ar - h) / 2.0)
border_x = 0
else:
border_x = border_y = out_border
# start_row, start_col = max(0, ymin - border_y), max(0, xmin - border_x)
# end_row, end_col = min(img_h - 1, ymax + border_y), min(img_w - 1, xmax + border_x)
start_row, start_col = ymin - border_y, xmin - border_x
end_row, end_col = ymax + border_y, xmax + border_x
if start_row < 0 or start_col < 0 or end_row >= img_h or end_col >= img_w:
msg = 'Invalid border {} for box {} in image {} of size {}'.format(
[border_x, border_y], [xmin, ymin, xmax, ymax], filename,
[img_w, img_h])
if allow_skipping_images:
print('\n' + msg + '\n')
continue
else:
raise AssertionError(msg)
if mask:
n_masks += 1
w, h = end_col - start_col, end_row - start_row
patch_img = img[start_row:end_row, start_col:end_col, :]
if not disable_resizing:
if max_dim > 0:
if w > h:
out_w = max_dim
out_h = 0
else:
out_h = max_dim
out_w = 0
elif min_dim > 0:
if w < h:
out_w = min_dim
out_h = 0
else:
out_h = min_dim
out_w = 0
else:
out_w, out_h = out_mask_size
scale_x = float(out_w) / float(w)
scale_y = float(out_h) / float(h)
if scale_x == 0:
scale_x = scale_y
out_w = int(w * scale_x)
elif scale_y == 0:
scale_y = scale_x
out_h = int(h * scale_y)
try:
patch_img = cv2.resize(patch_img, (out_w, out_h))
# print('patch_img: {}'.format(patch_img.shape))
except cv2.error as e:
print('patch_img: {}'.format(patch_img.shape))
print('out_size: {}, {}'.format(start_row, start_col))
print('out_size: {}, {}'.format(end_row, end_col))
print('out_size: {}, {}'.format(out_w, out_h))
raise cv2.error(e)
else:
out_w, out_h = w, h
_label = label
if id_number is None:
id_number = -1
if id_number > 0:
_label = '{}_{}'.format(_label, id_number)
if enable_out_suffix:
out_fname = '{}_{}_{}'.format(filename_no_ext, obj_id, label)
else:
out_fname = filename_no_ext
_xmin, _ymin = int((xmin - start_col) * scale_x), int(
(ymin - start_row) * scale_y)
_xmax, _ymax = int((xmax - start_col) * scale_x), int(
(ymax - start_row) * scale_y)
if map_to_bbox:
if not img_written:
img_written = 1
out_img_path = os.path.join(img_root_path, filename)
cv2.imwrite(out_img_path, img)
if enable_xml_annotations:
imageShape = [xml_reader.height, xml_reader.width, 3]
xml_writer = PascalVocWriter(out_xml_path, filename,
imageShape)
if mask:
if enable_xml_annotations:
bndbox = [xmin, ymin, xmax, ymax]
xml_writer.addBndBox(bndbox[0], bndbox[1], bndbox[2],
bndbox[3], label, difficult,
bbox_source, id_number, score,
mask, mask_img)
raw_data = {
'target_id': int(id_number),
'filename': filename,
'width': img_w,
'height': img_h,
'class': label,
'xmin': xmin,
'ymin': ymin,
'xmax': xmax,
'ymax': ymax
}
if show_img:
cv2.rectangle(img, (xmin, ymin), (xmax, ymax), (0, 255, 0),
2)
disp_img = img
else:
img_out_fname = out_fname + '.jpg'
if mask:
out_img_path = os.path.join(img_root_path, img_out_fname)
cv2.imwrite(out_img_path, patch_img)
if enable_xml_annotations:
n_mask = len(mask)
_mask = []
for i in range(n_mask):
_mask.append([(mask[i][0] - start_col) * scale_x,
(mask[i][1] - start_row) * scale_y,
mask[i][2]])
imageShape = [xml_reader.height, xml_reader.width, 3]
xml_writer = PascalVocWriter(out_xml_path,
xml_reader.filename,
imageShape)
bndbox = [_xmin, _ymin, _xmax, _ymax]
xml_writer.addBndBox(_xmin, _ymin, _xmax, _ymax, label,
difficult, bbox_source, id_number,
score, _mask)
raw_data = {
'target_id': int(id_number),
'filename': img_out_fname,
'width': out_w,
'height': out_h,
'class': label,
'xmin': _xmin,
'ymin': _ymin,
'xmax': _xmax,
'ymax': _ymax
}
if show_img:
cv2.rectangle(patch_img, (_xmin, _ymin), (_xmax, _ymax),
(0, 255, 0), 2)
disp_img = patch_img
if mask:
if mask_img is None:
mask_img = np.zeros_like(img)
# print('border_x: {}'.format(border_x))
# print('border_y: {}'.format(border_y))
# print('scale_x: {}'.format(scale_x))
# print('scale_y: {}'.format(scale_y))
#
# print('xmin: {}'.format(xmin))
# print('ymin: {}'.format(ymin))
mask_pts = [[(x - xmin + border_x) * scale_x,
(y - ymin + border_y) * scale_y]
for x, y in mask_pts]
curr_mask = np.zeros_like(patch_img, dtype=np.uint8)
# print('mask_img: {}'.format(mask_img.shape))
mask_out_fname = out_fname + '.png'
# np.savetxt('mask_seq_mask_pts.txt', mask_pts, fmt='%.6f')
curr_mask = cv2.fillPoly(
curr_mask, np.array([
mask_pts,
], dtype=np.int32), mask_pix_val)
# print('min: {} max: {}'.format(
# np.min(mask_img.flatten()),
# np.max(mask_img.flatten()))
# )
if map_to_bbox:
mask_img = map_mask_to_bbox(
(xmin, ymin, xmax, ymax), curr_mask, fixed_ar,
out_border, mask_img.shape, mask_img)
else:
mask_img = curr_mask
out_mask_path = os.path.join(out_mask_root_path,
mask_out_fname)
cv2.imwrite(out_mask_path, mask_img)
_train_fnames.append((out_img_path, out_mask_path))
if show_img:
disp_mask_img = mask_img.copy()
if save_raw_mask:
disp_mask_img[disp_mask_img > 0] = 255
blended_img = np.asarray(
Image.blend(Image.fromarray(patch_img),
Image.fromarray(disp_mask_img), 0.5))
disp_img = np.concatenate(
(disp_img, disp_mask_img, blended_img), axis=1)
csv_raw.append(raw_data)
else:
test_csv_raw.append(raw_data)
if not map_to_bbox:
_test_fnames.append(out_img_path)
if show_img and not map_to_bbox:
# if _pause:
# print('frame {} :: {}'.format(file_id, filename))
cv2.imshow(win_name, disp_img)
k = cv2.waitKey(1 - _pause)
if k == ord('q'):
_exit = 1
break
elif k == 27:
_exit_seq = 1
break
elif k == 32:
_pause = 1 - _pause
obj_id += 1
if map_to_bbox and img is not None:
out_img_path = os.path.join(out_img_root_path, filename)
if save_train and mask_img is not None:
mask_out_fname = filename_no_ext + '.png'
out_mask_path = os.path.join(out_mask_root_path,
mask_out_fname)
cv2.imwrite(out_mask_path, mask_img)
if enable_xml_annotations:
out_xml_file = os.path.join(out_xml_path,
os.path.basename(file))
xml_writer.save(targetFile=out_xml_file)
_train_fnames.append((out_img_path, out_mask_path))
if show_img:
disp_mask_img = mask_img
if save_raw_mask:
disp_mask_img[disp_mask_img > 0] = 255
blended_img = np.asarray(
Image.blend(Image.fromarray(img),
Image.fromarray(disp_mask_img), 0.5))
disp_img = np.concatenate(
(disp_img, disp_mask_img, blended_img), axis=1)
elif save_test:
out_img_path = os.path.join(test_img_root_path, filename)
if out_img_path in _test_fnames:
raise IOError(
'Duplicate out_img_path: {}'.format(out_img_path))
_test_fnames.append(out_img_path)
if show_img and disp_img is not None:
cv2.imshow(win_name, disp_img)
k = cv2.waitKey(1 - _pause)
if k == ord('q'):
_exit = 1
break
elif k == 27:
break
elif k == 32:
_pause = 1 - _pause
if _exit:
break
sys.stdout.write(
'\rDone {:d}/{:d} files {:s} ({:d} masks found)'.format(
file_id + 1, n_files, filename, n_masks))
sys.stdout.flush()
if masks_per_seq > 0 and n_masks >= masks_per_seq:
break
sys.stdout.write('\n')
sys.stdout.flush()
if not _exit_seq and save_train and n_masks == 0:
raise IOError('\nNo masks found for {}\n'.format(seq_path))
train_csv_path = test_csv_path = ''
if csv_raw:
print('Saved {} labeled files in training sequence'.format(
len(csv_raw)))
train_csv_path = os.path.join(out_img_root_path, 'annotations.csv')
pd.DataFrame(csv_raw).to_csv(train_csv_path)
if test_csv_raw:
print('Saved {} unlabeled files in test sequence'.format(
len(test_csv_raw)))
test_csv_path = os.path.join(test_img_root_path, 'annotations.csv')
pd.DataFrame(test_csv_raw).to_csv(test_csv_path)
if show_img:
cv2.destroyWindow(win_name)
if save_train and train_fnames is not None:
train_fnames[
out_seq_name] = _train_fnames, train_root_path, csv_raw, train_csv_path
if save_test and test_fnames is not None:
test_fnames[
out_test_seq_name] = _test_fnames, test_img_root_path, test_csv_raw, test_csv_path
return n_masks
def visualize(vis_params,
logger,
img_path,
csv_path,
class_dict,
init_frame_id=0,
n_frames=0,
request_roi=None,
generator_mode=0,
enable_masks=0,
label='',
only_boxes=0,
crop_size=()):
"""
:param vis_params:
:param logger:
:param img_path:
:param csv_path:
:param class_dict:
:param init_frame_id:
:param n_frames:
:param request_roi:
:param generator_mode:
:return:
"""
global _pause, _quit
save_fname_templ = os.path.splitext(os.path.basename(img_path))[0]
# csv_path = os.path.join(img_path, 'annotations.csv')
df = pd.read_csv(csv_path)
frames_reader = get_frames_reader(img_path, save_as_bin=False)
if request_roi is not None:
frames_reader.setROI(request_roi)
class_labels = dict((v, k) for k, v in class_dict.items())
if generator_mode:
vis_params.show = 0
vis_params.save = 0
visualizer = Visualizer(vis_params, logger, class_labels)
init_frame = frames_reader.get_frame(init_frame_id)
height, width, _ = init_frame.shape
frame_size = width, height
visualizer.initialize(save_fname_templ, frame_size, _pause)
if n_frames <= 0:
n_frames = frames_reader.num_frames
print('Reading {:d} images from {:s}...'.format(n_frames, img_path))
for frame_id in range(init_frame_id, n_frames):
try:
curr_frame = frames_reader.get_frame(frame_id)
except IOError as e:
print('{}'.format(e))
break
if only_boxes:
curr_frame = np.zeros_like(curr_frame)
file_path = frames_reader.get_file_path()
if file_path is None:
print('Visualization is only supported on image sequence data')
return
filename = os.path.basename(file_path)
multiple_instance = df.loc[df['filename'] == filename]
# Total # of object instances in a file
n_bboxes = len(multiple_instance.index)
# Remove from df (avoids duplication)
df = df.drop(multiple_instance.index[:n_bboxes])
frame_data = []
masks = []
generic_target_id = -1
if enable_masks:
filename = os.path.basename(file_path)
xml_path = os.path.join(img_path, 'annotations',
os.path.splitext(filename)[0] + '.xml')
if not os.path.isfile(xml_path):
print('{} :: annotations xml file not found: {}'.format(
filename, xml_path))
continue
xml_reader = PascalVocReader(xml_path)
shapes = xml_reader.getShapes()
n_shapes = len(shapes)
if n_shapes != n_bboxes:
raise IOError(
'Mismatch between n_bboxes in xml: {} and csv: {}'.format(
n_shapes, n_bboxes))
for box_id in range(n_bboxes):
bbox = multiple_instance.iloc[box_id]
try:
target_id = bbox['target_id']
except KeyError:
target_id = generic_target_id
generic_target_id -= 1
xmin = bbox.loc['xmin']
ymin = bbox.loc['ymin']
xmax = bbox.loc['xmax']
ymax = bbox.loc['ymax']
class_name = bbox.loc['class']
try:
class_id = class_dict[str(class_name)]
except KeyError:
print('Ignoring annotation with invalid class: {}'.format(
class_name))
continue
width = xmax - xmin
height = ymax - ymin
curr_frame_data = [
frame_id, target_id, xmin, ymin, width, height, class_id
]
if enable_masks:
mask = shapes[box_id][-2]
if mask is not None:
_contour_pts = Shape.getContourPts(mask)
masks.append(_contour_pts)
frame_data.append(curr_frame_data)
frame_data = np.asarray(frame_data)
res = visualizer.update(frame_id, curr_frame, frame_data, masks, label,
crop_size)
if generator_mode:
yield res
# elif not res:
# break
_quit = visualizer._quit
_pause = visualizer._pause
visualizer.close()
frames_reader.close()
def visualize(self, request):
request_path = request["path"]
csv_path = request["csv_path"]
class_dict = request["class_dict"]
request_roi = request["roi"]
init_frame_id = request["frame_number"]
save_fname_templ = os.path.splitext(os.path.basename(request_path))[0]
df = pd.read_csv(csv_path)
if request_path != self.current_path:
self.frames_reader = get_frames_reader(
request_path, save_as_bin=self.params.save_as_bin)
if request_roi is not None:
self.frames_reader.setROI(request_roi)
self.current_path = request_path
class_labels = dict((v, k) for k, v in class_dict.items())
# print('self.params.visualizer.save: ', self.params.visualizer.save)
visualizer = Visualizer(self.params.visualizer, self.logger,
class_labels)
init_frame = self.frames_reader.get_frame(init_frame_id)
height, width, _ = init_frame.shape
frame_size = width, height
visualizer.initialize(save_fname_templ, frame_size)
n_frames = self.frames_reader.num_frames
for frame_id in range(init_frame_id, n_frames):
try:
curr_frame = self.frames_reader.get_frame(frame_id)
except IOError as e:
print('{}'.format(e))
break
file_path = self.frames_reader.get_file_path()
if file_path is None:
print('Visualization is only supported on image sequence data')
return
filename = os.path.basename(file_path)
multiple_instance = df.loc[df['filename'] == filename]
# Total # of object instances in a file
no_instances = len(multiple_instance.index)
# Remove from df (avoids duplication)
df = df.drop(multiple_instance.index[:no_instances])
frame_data = []
for instance in range(0, len(multiple_instance.index)):
target_id = multiple_instance.iloc[instance].loc['target_id']
xmin = multiple_instance.iloc[instance].loc['xmin']
ymin = multiple_instance.iloc[instance].loc['ymin']
xmax = multiple_instance.iloc[instance].loc['xmax']
ymax = multiple_instance.iloc[instance].loc['ymax']
class_name = multiple_instance.iloc[instance].loc['class']
class_id = class_dict[class_name]
width = xmax - xmin
height = ymax - ymin
frame_data.append(
[frame_id, target_id, xmin, ymin, width, height, class_id])
frame_data = np.asarray(frame_data)
if not visualizer.update(frame_id, curr_frame, frame_data):
break
visualizer.close()
def patchTracking(self, request=None, img_path=''):
if self.params.mode == 2:
sys.stdout.write('@@@ Starting tracker\n')
sys.stdout.flush()
cmd_args = sys.argv[1:]
else:
if request is not None:
cmd_args = request['cmd_args']
else:
cmd_args = ''
self.parseParams(self.parser, cmd_args)
if request is not None:
request_path = request["path"]
request_roi = request["roi"]
id_number = request['id_number']
init_frame_id = request["frame_number"]
init_bbox = request["bbox"]
init_bbox_list = [
int(init_bbox['xmin']),
int(init_bbox['ymin']),
int(init_bbox['xmax']),
int(init_bbox['ymax']),
]
label = request['label']
request_port = request["port"]
else:
request_path = img_path if img_path else self.params.img_path
request_roi = str2list(self.params.roi)
id_number = self.params.id_number
init_frame_id = self.params.init_frame_id
if self.params.init_bbox:
init_bbox_list = str2list(self.params.init_bbox)
init_bbox = {
'xmin': init_bbox_list[0],
'ymin': init_bbox_list[1],
'xmax': init_bbox_list[2],
'ymax': init_bbox_list[3],
}
else:
init_bbox = {}
label = request_port = None
gt_available = 0
if request_path != self.current_path:
self.frames_reader = get_frames_reader(request_path, save_as_bin=self.params.save_as_bin)
if request_roi is not None:
self.frames_reader.setROI(request_roi)
self.current_path = request_path
if not init_bbox:
csv_path = os.path.join(request_path, 'annotations.csv')
print('Reading annotations from {}'.format(csv_path))
import pandas as pd
df_gt = pd.read_csv(csv_path)
_ = self.frames_reader.get_frame(init_frame_id)
file_path = self.frames_reader.get_file_path()
filename = os.path.basename(file_path)
multiple_instance = df_gt.loc[df_gt['filename'] == filename]
bbox = multiple_instance.iloc[0]
xmin = bbox.loc['xmin']
ymin = bbox.loc['ymin']
xmax = bbox.loc['xmax']
ymax = bbox.loc['ymax']
init_bbox = {
'xmin': xmin,
'ymin': ymin,
'xmax': xmax,
'ymax': ymax,
}
gt_available = 1
show_only = (self.params.mode == 1)
tracker = PatchTracker(self.params.patch_tracker, self.logger, id_number, label, show_only=show_only)
if not tracker.is_created:
return
init_frame = self.frames_reader.get_frame(init_frame_id)
tracker.initialize(init_frame, init_bbox)
if not tracker.is_initialized:
self.logger.error('Tracker initialization was unsuccessful')
return
n_frames = self.frames_reader.num_frames
if self.params.end_frame_id >= init_frame_id:
end_frame_id = self.params.end_frame_id
else:
end_frame_id = n_frames - 1
if self.params.mode == 1:
if self.client is None:
self.connectToExecutionServer()
# print('init_bbox_list: ', init_bbox_list)
remote_bbox = list2str(init_bbox_list)
remote_img_path = os.path.join(self.params.remote_img_root_path, os.path.basename(request_path))
cd_command = 'cd {:s}'.format(self.params.remote_path)
exec_command = 'python3 Server.py --mode=2 ' \
'--cfg={:s} --img_path={:s} --id_number={:d} ' \
'--init_frame_id={:d} --init_bbox={:s}' \
' --patch_tracker.tracker_type={:d}' \
' --patch_tracker.cv_tracker_type={:d}' \
' --patch_tracker.show=0' \
'' \
'\n'.format(
self.params.remote_cfg,
remote_img_path,
id_number,
init_frame_id,
remote_bbox,
self.params.patch_tracker.tracker_type,
self.params.patch_tracker.cv_tracker_type,
)
# command = '{} && {}'.format(cd_command, exec_command)
# if request_roi is not None:
# remote_roi = list2str(request_roi)
# command = '{:s} {:s}'.format(command, remote_roi)
#
print('Running:\n{:s}'.format(exec_command))
curr_corners = np.zeros((2, 4), dtype=np.float64)
# channel = self.client.invoke_shell(width=1000, height=3000)
# _stdout = channel.makefile()
self.channel.send(cd_command + '\n')
# channel.send("sudo -s\n'''\n")
# channel.send("'''" + '\n')
self.channel.send(exec_command + '\n')
# channel.send('exit' + '\n')
# s = channel.recv(4096)
# print(s)
# client.close()
# sys.exit()
# _stdin, _stdout, _stderr = client.exec_command(command)
# channel = _stdout.channel
# print(_stdout.readlines())
# while not _stderr.channel.recv_exit_status():
# if _stderr.channel.recv_ready():
# print(_stderr.read())
pid = None
tracking_started = 0
wait_start_time = tracking_start_time = time.clock()
while not self.channel.exit_status_ready():
if not self.channel.recv_ready():
wait_time = time.clock() - wait_start_time
# sys.stdout.write('Waiting for stdout for {:f} secs\n'.format(wait_time))
# sys.stdout.flush()
if wait_time > self.params.wait_timeout:
print('Waiting time threshold exceeded')
break
continue
wait_start_time = time.clock()
# sys.stdout.write('\n')
# sys.stdout.flush()
# remote_output = channel.recv(45).decode("utf-8")
remote_output = self._stdout.readline().replace("^C", "")
# print('remote_output: ', remote_output)
if remote_output.startswith('@@@'):
tracking_started = 1
continue
if not remote_output.startswith('###'):
sys.stdout.write(remote_output)
sys.stdout.flush()
continue
if not tracking_started:
continue
# sys.stdout.write(remote_output)
# sys.stdout.flush()
result_list = remote_output.strip().split()
# print('remote_output: ', remote_output)
# print('result_list: ', result_list)
if len(result_list) != 8:
print('remote_output: ', remote_output)
print('result_list: ', result_list)
raise SystemError('Invalid output from the remote server')
pid = int(result_list[1])
frame_id = int(result_list[2])
xmin = int(result_list[3])
ymin = int(result_list[4])
xmax = int(result_list[5])
ymax = int(result_list[6])
remote_fps = float(result_list[7])
curr_corners[:, 0] = (xmin, ymin)
curr_corners[:, 1] = (xmax, ymin)
curr_corners[:, 2] = (xmax, ymax)
curr_corners[:, 3] = (xmin, ymax)
try:
curr_frame = self.frames_reader.get_frame(frame_id)
except IOError as e:
print('{}'.format(e))
break
end_time = time.clock()
fps = 1.0 / (end_time - tracking_start_time)
tracker.show(curr_frame, curr_corners, frame_id, fps, remote_fps)
out_bbox = dict(
xmin=xmin,
ymin=ymin,
xmax=xmax,
ymax=ymax,
)
self.send(curr_frame, out_bbox, label, request_path, frame_id, id_number, request_port)
tracking_start_time = end_time
if frame_id > end_frame_id or tracker.is_terminated:
# exit_client = paramiko.SSHClient()
# exit_client.set_missing_host_key_policy(paramiko.AutoAddPolicy)
# exit_client.connect(self.params.hostname, username=self.params.username,
# password=self.params.password)
# _stdin, _stdout, _stderr = exit_client.exec_command()
# exit_client.close()
# self.channel.send(chr(3))
# self.channel.send(chr(3))
# self.channel.send(chr(3))
# self.channel.send('exit' + '\n')
# channel_closed = 1
# channel.close()
# self.channel.send('pkill -P {:d}\n'.format(pid))
break
# while not channel.recv_ready():
# sys.stdout.write('\rwaiting')
# sys.stdout.flush()
# continue
tracker.close()
self.channel.send(chr(3))
if pid is not None:
self.channel.send('pkill -P {:d}\n'.format(pid))
# self.flushChannel()
# p = multiprocessing.Process(target=self.flushChannel)
# p.start()
# # Wait for 1 second or until process finishes
# p.join(self.params.flush_timeout)
# if p.is_alive():
# p.terminate()
# p.join()
print('\ndone execution')
# _stdout.close()
# _stderr.close()
# channel.send('exit' + '\n')
# self.channel.close()
# client.close()
return
# init_frame = self.frames_reader.get_frame(init_frame_id)
# init_file_path = self.frames_reader.get_file_path()
# try:
# self.runTracker(tracker, init_frame_id)
# except KeyboardInterrupt:
# pass
# self.logger.info('Tracking target {:d} in sequence with {:d} frames '
# 'starting from frame {:d}'.format(
# id_numbers[0], n_frames, init_frame_id + 1))
save_path = ''
if self.params.save_dir:
file_path = self.frames_reader.get_file_path()
save_path = os.path.join('log', self.params.save_dir, os.path.basename(os.path.dirname(file_path)))
if not os.path.isdir(save_path):
os.makedirs(save_path)
save_csv_path = os.path.join(save_path, 'annotations.csv')
print('Saving results csv to {}'.format(save_csv_path))
if self.params.track_init_frame:
start_frame_id = init_frame_id
else:
start_frame_id = init_frame_id + 1
csv_raw = []
if label is None:
label = 'generic'
for frame_id in range(start_frame_id, end_frame_id + 1):
try:
curr_frame = self.frames_reader.get_frame(frame_id)
except IOError as e:
print('{}'.format(e))
break
file_path = self.frames_reader.get_file_path()
filename = os.path.basename(file_path)
gt_bbox = None
if gt_available:
multiple_instance = df_gt.loc[df_gt['filename'] == filename]
bbox = multiple_instance.iloc[0]
xmin = bbox.loc['xmin']
ymin = bbox.loc['ymin']
xmax = bbox.loc['xmax']
ymax = bbox.loc['ymax']
label = bbox.loc['class']
gt_bbox = {
'xmin': xmin,
'ymin': ymin,
'xmax': xmax,
'ymax': ymax,
}
try:
fps = tracker.update(curr_frame, frame_id, gt_bbox=gt_bbox)
except KeyboardInterrupt:
break
if tracker.out_bbox is None:
# self.logger.error('Tracker update was unsuccessful')
break
if save_path:
if tracker.curr_mask_cropped is not None:
mask_filename = os.path.splitext(filename)[0] + '.png'
save_mask_path = os.path.join(save_path, mask_filename)
curr_mask_norm = (tracker.curr_mask_cropped * 255.0).astype(np.uint8)
cv2.imwrite(save_mask_path, curr_mask_norm)
mask_filename_bin = os.path.splitext(filename)[0] + '.npy'
save_mask_path_bin = os.path.join(save_path, mask_filename_bin)
np.save(save_mask_path_bin, tracker.curr_mask_cropped)
if self.params.save_csv:
orig_height, orig_width = curr_frame.shape[:2]
xmin = tracker.out_bbox['xmin']
xmax = tracker.out_bbox['xmax']
ymin = tracker.out_bbox['ymin']
ymax = tracker.out_bbox['ymax']
raw_data = {
'filename': filename,
'width': orig_width,
'height': orig_height,
'class': label,
'xmin': int(xmin),
'ymin': int(ymin),
'xmax': int(xmax),
'ymax': int(ymax),
'confidence': tracker.score
}
csv_raw.append(raw_data)
if self.params.mode == 2 and not self.params.patch_tracker.show:
sys.stdout.write('### {:d} {:d} {:d} {:d} {:d} {:d} {:5.2f}\n'.format(
self.pid, frame_id, tracker.out_bbox['xmin'], tracker.out_bbox['ymin'],
tracker.out_bbox['xmax'], tracker.out_bbox['ymax'], fps
))
sys.stdout.flush()
continue
if request_port is not None:
mask = os.path.abspath(save_mask_path_bin)
# if tracker.curr_mask_cropped is not None:
# # mask = np.expand_dims(tracker.curr_mask, axis=0).tolist()
# mask = tracker.curr_mask_cropped.tolist()
# else:
# mask = None
self.send(curr_frame, tracker.out_bbox, label, request_path, frame_id,
id_number, request_port, masks=mask)
# self.single_object_tracking_results.append(tracking_result)
if tracker.is_terminated:
break
sys.stdout.write('Closing tracker...\n')
sys.stdout.flush()
tracker.close()
if save_path and self.params.save_csv:
df = pd.DataFrame(csv_raw)
df.to_csv(save_csv_path)
