def openpose_convert_labelme_format(input_json, input_image_path=None, output_dir=None):
openPose_object = OpenPose_JSON(input_json)
labelme_object = Labelme_JSON()
image_name = utils_io_folder.get_file_name_from_path(input_image_path) # like 00000001.jpg
labelme_object.imagePath = input_image_path
labelme_object.imageData = utils_image.img_encode_to_base64(labelme_object.imagePath)
image_arr = utils_image.img_b64_to_arr(labelme_object.imageData)
labelme_object.imageHeight, labelme_object.imageWidth = image_arr.shape[0], image_arr.shape[1]
for person in openPose_object.people:
track_id = person["person_id"]
pose_keypoints_2d = person["pose_keypoints_2d"]
hand_left_keypoints_2d = person["hand_left_keypoints_2d"]
hand_right_keypoints_2d = person["hand_right_keypoints_2d"]
for index, keypoint in enumerate(pose_keypoints_2d):
labelme_object.shapes.append(point_shape(track_id, pose_keypoints_order[index], keypoint))
for index, keypoint in enumerate(hand_left_keypoints_2d):
labelme_object.shapes.append(point_shape(track_id, hand_keypoints_order[index].format("left_"), keypoint))
for index, keypoint in enumerate(hand_right_keypoints_2d):
labelme_object.shapes.append(point_shape(track_id, hand_keypoints_order[index].format("right_"), keypoint))
if output_dir is None:
# 输出 默认在input_json的路径下创建labelme文件夹。
output_dir = os.path.join(utils_io_folder.get_parent_folder_from_path(input_json))
# utils_io_folder.create_folder(output_dir)
output_path = os.path.join(output_dir, utils_io_folder.replace_file_ext(image_name, "json"))
else:
utils_io_folder.create_folder(output_dir)
output_path = os.path.join(output_dir, utils_io_folder.replace_file_ext(image_name, "json"))
utils_json.write_json_to_file(labelme_object.to_dict(), output_path)
def openSVAI_convert_labelme_format(input_json, input_images_dir=None, output_dir=None):
openSVAI_object = OpenSVAI_JSON(input_json)
for frame in openSVAI_object.frames:
labelme_object = Labelme_JSON()
image_info = frame["image"]
candidates_info = frame["candidates"]
image_name = image_info["name"] # like 00000001.jpg
if input_images_dir is None:
labelme_object.imagePath = os.path.join(image_info["folder"], image_info["name"])
else:
labelme_object.imagePath = os.path.join(input_images_dir, image_name)
labelme_object.imageData = utils_image.img_encode_to_base64(labelme_object.imagePath)
image_arr = utils_image.img_b64_to_arr(labelme_object.imageData)
labelme_object.imageHeight, labelme_object.imageWidth = image_arr.shape[0], image_arr.shape[1]
for candidate in candidates_info:
track_id = candidate['track_id'] # 对应 labelme的group_id
det_bbox = candidate["det_bbox"]
# det_bbox (open_SVAI) - rectangle (labelme)
labelme_object.shapes.append(rectangle_shape(track_id, det_bbox))
keypoints = numpy.array(candidate['pose_keypoints_2d']).reshape(-1, 3)
for index, keypoint in enumerate(keypoints):
# keypoint (open_SVAI) - point (labelme)
labelme_object.shapes.append(point_shape(track_id, keypoints_order[index], keypoint))
if output_dir is None:
# 输出 默认在input_json的路径下创建labelme文件夹。
output_dir = os.path.join(utils_io_folder.get_parent_folder_from_path(input_json), "labelme")
utils_io_folder.create_folder(output_dir)
output_path = os.path.join(output_dir, utils_io_folder.replace_file_ext(image_name, "json"))
else:
utils_io_folder.create_folder(output_dir)
output_path = os.path.join(output_dir, utils_io_folder.replace_file_ext(image_name, "json"))
utils_json.write_json_to_file(labelme_object.to_dict(), output_path)
def start_convert():
labelme_base_path = "F:/DataSet/douyin/my_works/{}"
index = ["000003"]
for i in tqdm(index):
path = labelme_base_path.format(i)
print(path)
labelme_json_paths = utils_io_folder.get_immediate_childfile_paths(
path, ext="json")
# TODO 之后要改成在统一的一个目录下
output_folder = "F:/DataSet/douyin/my_works/{}".format(i)
utils_io_folder.create_folder(output_folder)
for labelme_json_path in tqdm(labelme_json_paths):
number = int(
utils_io_folder.get_file_name_without_ext_from_path(
labelme_json_path))
labelme_to_openpose(labelme_json_path, output_folder)
def start_tmp():
labelme_base_path = "E:/temp/anno/{}"
output_base_path = "E:/temp/anno/output/{}"
temp_video_types = ["test", "train"]
for video_type in temp_video_types:
openpose_json_folder_type = output_base_path.format(video_type)
labelme_folder = labelme_base_path.format(video_type)
folder_name = utils_io_folder.get_file_name_without_ext_from_path(labelme_folder)
openpose_json_folder = os.path.join(openpose_json_folder_type, folder_name)
utils_io_folder.create_folder(openpose_json_folder)
labelme_json_paths = utils_io_folder.get_immediate_childfile_paths(labelme_folder, ext="json")
for labelme_json_path in tqdm(labelme_json_paths):
# convert json one by one
labelme_to_openpose(labelme_json_path, openpose_json_folder)
def work():
root_path = "/media/jion/D/chenhaoming/DataSet/PoseTrack2017/cropped_image"
first_paths = utils_io_folder.get_immediate_subfolder_paths(root_path)
for first_path in first_paths:
second_paths = utils_io_folder.get_immediate_subfolder_paths(
first_path)
for second_path in second_paths:
image_paths = utils_io_folder.get_immediate_childfile_paths(
second_path)
if image_paths is None or len(image_paths) == 0:
os.rmdir(second_path)
else:
pa = "/media/jion/D/chenhaoming/experiment/hmr"
print(pa + second_path[61:])
# new_path=""
utils_io_folder.create_folder(pa + second_path[61:])
def start_convert():
labelme_base_path = "F:/DataSet/DouYin/annotation_labelme_image/{}"
output_base_path = "F:/DataSet/DouYin/annotation_openpose/{}"
for video_type in tqdm(video_types):
openpose_json_folder_type = output_base_path.format(video_type)
labelme_folder = labelme_base_path.format(video_type)
video_folder_sequence = utils_io_folder.get_immediate_subfolder_paths(labelme_folder)
for video_folder in tqdm(video_folder_sequence):
folder_name = utils_io_folder.get_file_name_without_ext_from_path(video_folder)
if folder_name.isdigit():
openpose_json_folder = os.path.join(openpose_json_folder_type, folder_name)
utils_io_folder.create_folder(openpose_json_folder)
labelme_json_paths = utils_io_folder.get_immediate_childfile_paths(video_folder, ext="json")
for labelme_json_path in tqdm(labelme_json_paths):
# convert json one by one
labelme_to_openpose(labelme_json_path, openpose_json_folder)
def work():
cmd = "mv {} {}"
root_path = "/media/jion/D/chenhaoming/DataSet/PoseTrack2017/posetrack_data/images"
change_path = "/media/jion/D/chenhaoming/DataSet/PoseTrack2017/cropped_image/{}/{}"
folder_paths = utils_io_folder.get_immediate_subfolder_paths(root_path)
for folder_path in tqdm(folder_paths):
first_folder_name = utils_io_folder.get_file_name_without_ext_from_path(folder_path)
# print(first_folder_name)
paths = utils_io_folder.get_immediate_subfolder_paths(folder_path)
for path in paths:
second_folder_name = utils_io_folder.get_file_name_without_ext_from_path(path)
# print(second_folder_name)
images_path = utils_io_folder.get_immediate_childfile_paths(path)
new_path = change_path.format(first_folder_name, second_folder_name)
utils_io_folder.create_folder(new_path)
for image_path in images_path:
image_name = utils_io_folder.get_file_name_without_ext_from_path(image_path)
if image_name.find("transform") >= 0:
cmd_i = cmd.format(image_path, os.path.join(new_path, image_name))
os.system(cmd_i)
def make_images_from_video(video_path, outimages_path=None):
cap = cv2.VideoCapture(video_path)
isOpened = cap.isOpened()
video_length = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
fps = cap.get(cv2.CAP_PROP_FPS) # 帧率<每秒中展示多少张图片>
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) # 获取宽度
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) # 获取宽度
i = 0
if outimages_path is not None:
if not utils_io_folder.folder_exists(outimages_path):
utils_io_folder.create_folder(outimages_path)
assert isOpened, "Can't find video"
for index in tqdm(range(video_length)):
(flag, data) = cap.read() # 读取每一张 flag<读取是否成功> frame<内容>
file_name = "{0:08d}.jpg".format(index + 1)
if outimages_path is not None:
file_path = os.path.join(outimages_path, file_name)
else:
utils_io_folder.create_folder("output")
file_path = os.path.join("output", file_name)
if flag: # 读取成功的话
# 写入文件，1 文件名 2 文件内容 3 质量设置
cv2.imwrite(file_path, data, [cv2.IMWRITE_JPEG_QUALITY, 100])
def posetrack2017_convert_labelme_format(input_json, input_image_path_base,
output_dir_base):
"""
输入是一段视频的json 仅一个json文件，输出为每个帧的json文件
:param input_json:
:param input_image_path_base:
:param output_dir_base:
:return:
"""
annolist = utils_json.read_json_from_file(input_json)["annolist"]
for annotation in annolist:
# 处理一张图片
# openPose_object = OpenPose_JSON(input_json)
input_image_path = os.path.join(input_image_path_base,
annotation["image"][0]["name"])
labelme_object = Labelme_JSON()
image_name = utils_io_folder.get_file_name_from_path(
input_image_path) # like 00000001.jpg
labelme_object.imagePath = input_image_path
labelme_object.imageData = utils_image.img_encode_to_base64(
labelme_object.imagePath)
image_arr = utils_image.img_b64_to_arr(labelme_object.imageData)
labelme_object.imageHeight, labelme_object.imageWidth = image_arr.shape[
0], image_arr.shape[1]
people = annotation["annorect"]
for person in people:
track_id = person["track_id"]
annopoints = person["annopoints"][0]["point"]
for point in annopoints:
point_id = point["id"][0]
x = point["x"][0]
y = point["y"][0]
is_visible = point["is_visible"][0]
keypoint = [x, y]
labelme_object.shapes.append(
point_shape(
track_id,
joint_names[point_id] + "_vis_{}".format(is_visible),
keypoint))
for point_1 in annopoints:
point_1_id = point_1["id"][0]
x_1 = point_1["x"][0]
y_1 = point_1["y"][0]
for point_2 in annopoints:
point_2_id = point_2["id"][0]
x_2 = point_2["x"][0]
y_2 = point_2["y"][0]
for joint_pair in joint_pairs:
point_1_name = joint_names[point_1_id]
point_2_name = joint_names[point_2_id]
if joint_pair[0] == point_1_name and joint_pair[
1] == point_2_name:
labelme_object.shapes.append(
line_shape(track_id,
point_1_name + point_2_name,
[x_1, y_1], [x_2, y_2]))
path = annotation["image"][0]["name"]
output_dir = output_dir_base + path[path.index("/"):path.rindex("/"
)]
utils_io_folder.create_folder(output_dir)
output_path = os.path.join(
output_dir,
utils_io_folder.replace_file_ext(image_name, "json"))
utils_json.write_json_to_file(labelme_object.to_dict(),
output_path)