def cli():
predict_parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
# Predict (3D pose from images)
##Here we can add arguments to pass to the software.
# Pifpaf
nets.cli(predict_parser)
decoder.cli(predict_parser,
force_complete_pose=True,
instance_threshold=0.15)
predict_parser.add_argument(
'--scale',
default=0.2,
type=float,
help='change the scale of the image to preprocess')
predict_parser.add_argument('--pose3d_model',
help='path of 3Dpose model to load',
default='data/models/pose3d-v0_8.pkl')
predict_parser.add_argument('--webcam',
help='streaming',
action='store_true')
args = predict_parser.parse_args()
return args
def cli():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
nets.cli(parser)
decoder.cli(parser, force_complete_pose=False, instance_threshold=0.05)
parser.add_argument('images', nargs='*', help='input images')
parser.add_argument(
'--glob', help='glob expression for input images (for many images)')
parser.add_argument('-o',
'--output-directory',
help=('Output directory. When using this option, make '
'sure input images have distinct file names.'))
parser.add_argument('--show',
default=False,
action='store_true',
help='show image of output overlay')
parser.add_argument('--output-types',
nargs='+',
default=['skeleton', 'json'],
help='what to output: skeleton, keypoints, json')
parser.add_argument('--loader-workers',
default=2,
type=int,
help='number of workers for data loading')
parser.add_argument('--disable-cuda',
action='store_true',
help='disable CUDA')
parser.add_argument('--figure-width',
default=10.0,
type=float,
help='figure width')
parser.add_argument('--dpi-factor',
default=1.0,
type=float,
help='increase dpi of output image by this factor')
args = parser.parse_args()
# glob
if args.glob:
args.images += glob.glob(args.glob)
if not args.images:
raise Exception("no image files given")
# add args.device
args.device = torch.device('cpu')
args.pin_memory = False
if not args.disable_cuda and torch.cuda.is_available():
args.device = torch.device('cuda')
args.pin_memory = True
return args
def main():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
nets.cli(parser)
decoder.cli(parser, force_complete_pose=False, instance_threshold=0.1, seed_threshold=0.5)
parser.add_argument('--no-colored-connections',
dest='colored_connections', default=True, action='store_false',
help='do not use colored connections to draw poses')
parser.add_argument('--disable_cuda', action='store_true', default=None,
help='disable CUDA')
args = parser.parse_args()
# add args.device
args.device = torch.device('cpu')
if not args.disable_cuda and torch.cuda.is_available():
print('************************ using gpu *****************************')
args.device = torch.device('cuda')
# load model
model, _ = nets.factory_from_args(args)
model = model.to(args.device)
processor = decoder.factory_from_args(args, model)
# own coco val json
f = open('/home/chenjia/pedestrian_det/chenjiaPifPaf/splitjson/test_5000.json')
js = ujson.load(f)
img_paths = js['images'] # len==5000, img的相对路径
img_path_root = '/data/nfs_share/public/zxyang/human_data/detection_data_cpu'
out_root = '/home/chenjia/pedestrian_det/openpifpaf/show_eval'
# random check pred result
for i in range(50):
print('*************** run the ', i + 1, 'image ******************')
ind = np.random.randint(0, 4999)
img_path = os.path.join(img_path_root, img_paths[ind]['file_name'])
img = cv2.imread(img_path)
img = cv2.resize(img, (683, 384))
image = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
processed_image_cpu = transforms.image_transform(image.copy()) # normalize
processed_image = processed_image_cpu.contiguous().to(args.device, non_blocking=True) # transpose 2,0,1
fields = processor.fields(torch.unsqueeze(processed_image, 0))[0]
keypoint_sets, _ = processor.keypoint_sets(fields)
# plot pred result
plot_points(image, keypoint_sets, COCO_PERSON_SKELETON)
cv2.imwrite(os.path.join(out_root, str(ind) + '.jpg'), image)
def cli(): # pylint: disable=too-many-statements,too-many-branches
parser = argparse.ArgumentParser(
prog='python3 -m openpifpaf.video',
description=__doc__,
formatter_class=CustomFormatter,
)
parser.add_argument(
'--version',
action='version',
version='OpenPifPaf {version}'.format(version=__version__))
network.cli(parser)
decoder.cli(parser,
force_complete_pose=True,
instance_threshold=0.1,
seed_threshold=0.5)
pifpaf.show.cli(parser)
parser.add_argument(
'--source',
default='0',
help='OpenCV source url. Integer for webcams. Supports rtmp streams.')
parser.add_argument('--video-output',
default=None,
nargs='?',
const=True,
help='video with drawn skeleton output file')
parser.add_argument('--start-frame', type=int, default=0)
parser.add_argument('--skip-frames', type=int, default=1)
parser.add_argument('--key-frequency-hz', type=int, default=10)
parser.add_argument('--max-frames', type=int)
# Set the threshold for something
# parser.add_argument('--confidence-threshold', type=float, default=0.1,
# help='Set the threshold needed for a face to be blurred')
group = parser.add_argument_group('logging')
group.add_argument('-q',
'--quiet',
default=False,
action='store_true',
help='only show warning messages or above')
group.add_argument('--debug',
default=False,
action='store_true',
help='print debug messages')
args = parser.parse_args()
return args
def cli():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
nets.cli(parser)
decoder.cli(parser,
force_complete_pose=False,
instance_threshold=0.1,
seed_threshold=0.5)
parser.add_argument('--no-colored-connections',
dest='colored_connections',
default=True,
action='store_false',
help='do not use colored connections to draw poses')
parser.add_argument('--disable-cuda',
action='store_true',
help='disable CUDA')
parser.add_argument(
'--source',
default='0',
help='OpenCV source url. Integer for webcams. Or ipwebcam streams.')
parser.add_argument('--scale',
default=0.5,
type=float,
help='input image scale factor')
args = parser.parse_args()
# check whether source should be an int
if len(args.source) == 1:
args.source = int(args.source)
# add args.device
args.device = torch.device('cpu')
if not args.disable_cuda and torch.cuda.is_available():
args.device = torch.device('cuda')
return args
def cli():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
# Subparser definition
subparsers = parser.add_subparsers(
help='Different parsers for main actions', dest='command')
predict_parser = subparsers.add_parser("predict")
prep_parser = subparsers.add_parser("prep")
training_parser = subparsers.add_parser("train")
eval_parser = subparsers.add_parser("eval")
# Preprocess input data
prep_parser.add_argument('--dir_ann',
help='directory of annotations of 2d joints',
required=True)
prep_parser.add_argument(
'--dataset',
help=
'datasets to preprocess: nuscenes, nuscenes_teaser, nuscenes_mini, kitti',
default='kitti')
prep_parser.add_argument('--dir_nuscenes',
help='directory of nuscenes devkit',
default='data/nuscenes/')
prep_parser.add_argument('--iou_min',
help='minimum iou to match ground truth',
type=float,
default=0.3)
prep_parser.add_argument('--variance', help='new', action='store_true')
prep_parser.add_argument('--activity', help='new', action='store_true')
prep_parser.add_argument('--monocular', help='new', action='store_true')
# Predict (2D pose and/or 3D location from images)
# General
predict_parser.add_argument('--mode',
help='pifpaf, mono, stereo',
default='stereo')
predict_parser.add_argument('images', nargs='*', help='input images')
predict_parser.add_argument(
'--glob', help='glob expression for input images (for many images)')
predict_parser.add_argument('-o',
'--output-directory',
help='Output directory')
predict_parser.add_argument(
'--output_types',
nargs='+',
default=['json'],
help='what to output: json keypoints skeleton for Pifpaf'
'json bird front combined for Monoloco')
predict_parser.add_argument('--show',
help='to show images',
action='store_true')
# Pifpaf
nets.cli(predict_parser)
decoder.cli(predict_parser,
force_complete_pose=True,
instance_threshold=0.15)
predict_parser.add_argument(
'--scale',
default=1.0,
type=float,
help='change the scale of the image to preprocess')
# Monoloco
predict_parser.add_argument('--model',
help='path of MonoLoco model to load',
required=True)
predict_parser.add_argument('--hidden_size',
type=int,
help='Number of hidden units in the model',
default=512)
predict_parser.add_argument(
'--path_gt',
help='path of json file with gt 3d localization',
default='data/arrays/names-kitti-200615-1022.json')
predict_parser.add_argument('--transform',
help='transformation for the pose',
default='None')
predict_parser.add_argument('--draw_box',
help='to draw box in the images',
action='store_true')
predict_parser.add_argument('--z_max',
type=int,
help='maximum meters distance for predictions',
default=22)
predict_parser.add_argument('--n_dropout',
type=int,
help='Epistemic uncertainty evaluation',
default=0)
predict_parser.add_argument('--dropout',
type=float,
help='dropout parameter',
default=0.2)
predict_parser.add_argument(
'--show_all',
help='only predict ground-truth matches or all',
action='store_true')
# Social distancing and social interactions
predict_parser.add_argument('--social', help='social', action='store_true')
predict_parser.add_argument('--activity',
help='activity',
action='store_true')
predict_parser.add_argument('--json_dir', help='for social')
predict_parser.add_argument('--threshold_prob',
type=float,
help='concordance for samples',
default=0.25)
predict_parser.add_argument('--threshold_dist',
type=float,
help='min distance of people',
default=2)
predict_parser.add_argument('--margin',
type=float,
help='conservative for noise in orientation',
default=1.5)
predict_parser.add_argument('--radii',
type=tuple,
help='o-space radii',
default=(0.25, 1, 2))
# Training
training_parser.add_argument(
'--joints',
help='Json file with input joints',
default='data/arrays/joints-nuscenes_teaser-190513-1846.json')
training_parser.add_argument('--save',
help='whether to not save model and log file',
action='store_true')
training_parser.add_argument('-e',
'--epochs',
type=int,
help='number of epochs to train for',
default=500)
training_parser.add_argument('--bs',
type=int,
default=512,
help='input batch size')
training_parser.add_argument('--monocular',
help='whether to train monoloco',
action='store_true')
training_parser.add_argument('--dropout',
type=float,
help='dropout. Default no dropout',
default=0.2)
training_parser.add_argument('--lr',
type=float,
help='learning rate',
default=0.001)
training_parser.add_argument('--sched_step',
type=float,
help='scheduler step time (epochs)',
default=30)
training_parser.add_argument('--sched_gamma',
type=float,
help='Scheduler multiplication every step',
default=0.98)
training_parser.add_argument('--hidden_size',
type=int,
help='Number of hidden units in the model',
default=1024)
training_parser.add_argument('--n_stage',
type=int,
help='Number of stages in the model',
default=3)
training_parser.add_argument('--hyp',
help='run hyperparameters tuning',
action='store_true')
training_parser.add_argument('--multiplier',
type=int,
help='Size of the grid of hyp search',
default=1)
training_parser.add_argument(
'--r_seed',
type=int,
help='specify the seed for training and hyp tuning',
default=1)
training_parser.add_argument('--activity', help='new', action='store_true')
# Evaluation
eval_parser.add_argument('--dataset',
help='datasets to evaluate, kitti or nuscenes',
default='kitti')
eval_parser.add_argument('--geometric',
help='to evaluate geometric distance',
action='store_true')
eval_parser.add_argument('--generate',
help='create txt files for KITTI evaluation',
action='store_true')
eval_parser.add_argument(
'--dir_ann',
help='directory of annotations of 2d joints (for KITTI evaluation)')
eval_parser.add_argument('--model', help='path of MonoLoco model to load')
eval_parser.add_argument(
'--joints',
help='Json file with input joints to evaluate (for nuScenes evaluation)'
)
eval_parser.add_argument('--n_dropout',
type=int,
help='Epistemic uncertainty evaluation',
default=0)
eval_parser.add_argument('--dropout',
type=float,
help='dropout. Default no dropout',
default=0.2)
eval_parser.add_argument('--hidden_size',
type=int,
help='Number of hidden units in the model',
default=1024)
eval_parser.add_argument('--n_stage',
type=int,
help='Number of stages in the model',
default=3)
eval_parser.add_argument('--show',
help='whether to show statistic graphs',
action='store_true')
eval_parser.add_argument('--save',
help='whether to save statistic graphs',
action='store_true')
eval_parser.add_argument('--verbose',
help='verbosity of statistics',
action='store_true')
eval_parser.add_argument('--monocular',
help='whether to train using the baseline',
action='store_true')
eval_parser.add_argument('--new', help='new', action='store_true')
eval_parser.add_argument('--variance',
help='evaluate keypoints variance',
action='store_true')
eval_parser.add_argument('--activity',
help='evaluate activities',
action='store_true')
eval_parser.add_argument(
'--net',
help='Choose network: monoloco, monoloco_p, monoloco_pp, monstereo')
args = parser.parse_args()
return args
def main():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
nets.cli(parser)
decoder.cli(parser, force_complete_pose=False, instance_threshold=0.05)
parser.add_argument('--no-colored-connections',
dest='colored_connections',
default=True,
action='store_false',
help='do not use colored connections to draw poses')
parser.add_argument('--disable-cuda',
action='store_true',
help='disable CUDA')
parser.add_argument(
'--source',
default=0,
help='OpenCV source url. Integer for webcams. Or ipwebcam streams.')
parser.add_argument('--scale',
default=0.1,
type=float,
help='input image scale factor')
args = parser.parse_args()
# check whether source should be an int
if len(args.source) == 1:
args.source = int(args.source)
# add args.device
args.device = torch.device('cpu')
if not args.disable_cuda and torch.cuda.is_available():
args.device = torch.device('cuda')
# load model
model, _ = nets.factory(args)
model = model.to(args.device)
processors = decoder.factory(args, model)
last_loop = time.time()
capture = cv2.VideoCapture(args.source)
visualizers = None
while True:
ret, image_original = capture.read()
print(ret)
image = cv2.resize(image_original, None, fx=args.scale, fy=args.scale)
print('resized image size: {}'.format(image.shape))
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
if visualizers is None:
visualizers = [Visualizer(p, args)(image) for p in processors]
for v in visualizers:
v.send(None)
start = time.time()
processed_image_cpu = transforms.image_transform(image.copy())
processed_image = processed_image_cpu.contiguous().to(
args.device, non_blocking=True)
print('preprocessing time', time.time() - start)
fields = processors[0].fields(torch.unsqueeze(processed_image, 0))[0]
for v in visualizers:
v.send((image, fields))
print('loop time = {:.3}s, FPS = {:.3}'.format(
time.time() - last_loop, 1.0 / (time.time() - last_loop)))
last_loop = time.time()
def cli():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
nets.cli(parser)
decoder.cli(parser, force_complete_pose=True)
encoder.cli(parser)
parser.add_argument('--output',
default=None,
help='output filename without file extension')
parser.add_argument('-n', default=0, type=int, help='number of batches')
parser.add_argument('--skip-n', default=0, type=int, help='skip n batches')
parser.add_argument('--dataset',
choices=('val', 'test', 'test-dev'),
default='val',
help='dataset to evaluate')
parser.add_argument('--min-ann',
default=0,
type=int,
help='minimum number of truth annotations')
parser.add_argument('--batch-size', default=1, type=int, help='batch size')
parser.add_argument('--long-edge',
default=641,
type=int,
help='long edge of input images')
parser.add_argument('--loader-workers',
default=None,
type=int,
help='number of workers for data loading')
parser.add_argument('--skip-existing',
default=False,
action='store_true',
help='skip if output eval file exists already')
parser.add_argument('--disable-cuda',
action='store_true',
help='disable CUDA')
parser.add_argument('--write-predictions',
default=False,
action='store_true',
help='write a json and a zip file of the predictions')
parser.add_argument('--all-images',
default=False,
action='store_true',
help='run over all images irrespective of catIds')
group = parser.add_argument_group('logging')
group.add_argument('--debug',
default=False,
action='store_true',
help='print debug messages')
args = parser.parse_args()
logging.basicConfig()
log_level = logging.INFO if not args.debug else logging.DEBUG
logging.getLogger('openpifpaf').setLevel(log_level)
LOG.setLevel(log_level)
if args.loader_workers is None:
args.loader_workers = args.batch_size
if args.dataset == 'val':
args.image_dir = IMAGE_DIR_VAL
args.annotation_file = ANNOTATIONS_VAL
elif args.dataset == 'test':
args.image_dir = IMAGE_DIR_TEST
args.annotation_file = ANNOTATIONS_TEST
elif args.dataset == 'test-dev':
args.image_dir = IMAGE_DIR_TEST
args.annotation_file = ANNOTATIONS_TESTDEV
else:
raise Exception
if args.dataset in ('test', 'test-dev'
) and not args.write_predictions and not args.debug:
raise Exception('have to use --write-predictions for this dataset')
if args.dataset in ('test',
'test-dev') and not args.all_images and not args.debug:
raise Exception('have to use --all-images for this dataset')
# add args.device
args.device = torch.device('cpu')
args.pin_memory = False
if not args.disable_cuda and torch.cuda.is_available():
args.device = torch.device('cuda')
args.pin_memory = True
# generate a default output filename
if args.output is None:
args.output = default_output_name(args)
return args
def cli():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
# Subparser definition
subparsers = parser.add_subparsers(
help='Different parsers for main actions', dest='command')
predict_parser = subparsers.add_parser("predict")
prep_parser = subparsers.add_parser("prep")
training_parser = subparsers.add_parser("train")
eval_parser = subparsers.add_parser("eval")
# Preprocess input data
prep_parser.add_argument('--dir_ann',
help='directory of annotations of 2d joints',
required=True)
prep_parser.add_argument(
'--dataset',
help=
'datasets to preprocess: nuscenes, nuscenes_teaser, nuscenes_mini, kitti',
default='nuscenes')
prep_parser.add_argument('--dir_nuscenes',
help='directory of nuscenes devkit',
default='data/nuscenes/')
prep_parser.add_argument('--iou_min',
help='minimum iou to match ground truth',
type=float,
default=0.3)
# Predict (2D pose and/or 3D location from images)
# General
predict_parser.add_argument('--networks',
nargs='+',
help='Run pifpaf and/or monoloco',
default=['monoloco'])
predict_parser.add_argument('images', nargs='*', help='input images')
predict_parser.add_argument(
'--glob', help='glob expression for input images (for many images)')
predict_parser.add_argument('-o',
'--output-directory',
help='Output directory')
predict_parser.add_argument(
'--output_types',
nargs='+',
default=['json'],
help='what to output: json keypoints skeleton for Pifpaf'
'json bird front combined for Monoloco')
predict_parser.add_argument('--show',
help='to show images',
action='store_true')
# Pifpaf
nets.cli(predict_parser)
decoder.cli(predict_parser,
force_complete_pose=True,
instance_threshold=0.1)
predict_parser.add_argument(
'--scale',
default=1.0,
type=float,
help='change the scale of the image to preprocess')
# Monoloco
predict_parser.add_argument('--model',
help='path of MonoLoco model to load',
default="data/models/monoloco-190513-1437.pkl")
predict_parser.add_argument('--hidden_size',
type=int,
help='Number of hidden units in the model',
default=256)
predict_parser.add_argument(
'--path_gt',
help='path of json file with gt 3d localization',
default='data/arrays/names-kitti-190710-1206.json')
predict_parser.add_argument('--transform',
help='transformation for the pose',
default='None')
predict_parser.add_argument('--draw_box',
help='to draw box in the images',
action='store_true')
predict_parser.add_argument('--predict',
help='whether to make prediction',
action='store_true')
predict_parser.add_argument('--z_max',
type=int,
help='maximum meters distance for predictions',
default=22)
predict_parser.add_argument('--n_dropout',
type=int,
help='Epistemic uncertainty evaluation',
default=0)
predict_parser.add_argument('--dropout',
type=float,
help='dropout parameter',
default=0.2)
predict_parser.add_argument('--webcam',
help='monoloco streaming',
action='store_true')
# Training
training_parser.add_argument(
'--joints',
help='Json file with input joints',
default='data/arrays/joints-nuscenes_teaser-190513-1846.json')
training_parser.add_argument('--save',
help='whether to not save model and log file',
action='store_false')
training_parser.add_argument('-e',
'--epochs',
type=int,
help='number of epochs to train for',
default=150)
training_parser.add_argument('--bs',
type=int,
default=256,
help='input batch size')
training_parser.add_argument('--baseline',
help='whether to train using the baseline',
action='store_true')
training_parser.add_argument('--dropout',
type=float,
help='dropout. Default no dropout',
default=0.2)
training_parser.add_argument('--lr',
type=float,
help='learning rate',
default=0.002)
training_parser.add_argument('--sched_step',
type=float,
help='scheduler step time (epochs)',
default=20)
training_parser.add_argument('--sched_gamma',
type=float,
help='Scheduler multiplication every step',
default=0.9)
training_parser.add_argument('--hidden_size',
type=int,
help='Number of hidden units in the model',
default=256)
training_parser.add_argument('--n_stage',
type=int,
help='Number of stages in the model',
default=3)
training_parser.add_argument('--hyp',
help='run hyperparameters tuning',
action='store_true')
training_parser.add_argument('--multiplier',
type=int,
help='Size of the grid of hyp search',
default=1)
training_parser.add_argument(
'--r_seed',
type=int,
help='specify the seed for training and hyp tuning',
default=1)
# Evaluation
eval_parser.add_argument('--dataset',
help='datasets to evaluate, kitti or nuscenes',
default='kitti')
eval_parser.add_argument('--geometric',
help='to evaluate geometric distance',
action='store_true')
eval_parser.add_argument('--generate',
help='create txt files for KITTI evaluation',
action='store_true')
eval_parser.add_argument(
'--dir_ann',
help='directory of annotations of 2d joints (for KITTI evaluation')
eval_parser.add_argument('--model',
help='path of MonoLoco model to load',
required=True)
eval_parser.add_argument(
'--joints',
help='Json file with input joints to evaluate (for nuScenes evaluation)'
)
eval_parser.add_argument('--n_dropout',
type=int,
help='Epistemic uncertainty evaluation',
default=0)
eval_parser.add_argument('--dropout',
type=float,
help='dropout. Default no dropout',
default=0.2)
eval_parser.add_argument('--hidden_size',
type=int,
help='Number of hidden units in the model',
default=256)
eval_parser.add_argument('--n_stage',
type=int,
help='Number of stages in the model',
default=3)
eval_parser.add_argument('--show',
help='whether to show statistic graphs',
action='store_true')
eval_parser.add_argument('--verbose',
help='verbosity of statistics',
action='store_true')
eval_parser.add_argument('--stereo',
help='include stereo baseline results',
action='store_true')
args = parser.parse_args()
return args
def main():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
nets.cli(parser)
decoder.cli(parser, force_complete_pose=False, instance_threshold=0.05)
parser.add_argument('--no-colored-connections',
dest='colored_connections', default=True, action='store_false',
help='do not use colored connections to draw poses')
parser.add_argument('--disable-cuda', action='store_true',
help='disable CUDA')
parser.add_argument('--source', default=0,
help='OpenCV source url. Integer for webcams. Or ipwebcam streams.')
parser.add_argument('--scale', default=0.1, type=float,
help='input image scale factor')
args = parser.parse_args()
# check whether source should be an int
if len(args.source) == 1:
args.source = int(args.source)
# add args.device
args.device = torch.device('cpu')
if not args.disable_cuda and torch.cuda.is_available():
args.device = torch.device('cuda')
# load model
model, _ = nets.factory(args)
model = model.to(args.device)
processors = decoder.factory(args, model)
print(args)
last_loop = time.time()
capture = cv2.VideoCapture(args.source)
visualizers = None
while True:
ret , image_original = capture.read()
print(ret)
image = cv2.resize(image_original, None, fx=args.scale, fy=args.scale)
print('resized image size: {}'.format(image.shape))
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
if visualizers is None:
visualizers = [Visualizer(p, args)(image) for p in processors]
for v in visualizers:
v.send(None)
start = time.time()
processed_image_cpu = transforms.image_transform(image.copy())
processed_image = processed_image_cpu.contiguous().to(args.device, non_blocking=True)
print('preprocessing time', time.time() - start)
fields = processors[0].fields(torch.unsqueeze(processed_image, 0))[0]
for v in visualizers:
v.send((image, fields))
print('loop time = {:.3}s, FPS = {:.3}'.format(
time.time() - last_loop, 1.0 / (time.time() - last_loop)))
last_loop = time.time()
def cli():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
# Subparser definition
subparsers = parser.add_subparsers(
help='Different parsers for main actions', dest='command')
predict_parser = subparsers.add_parser("predict")
prep_parser = subparsers.add_parser("prep")
training_parser = subparsers.add_parser("train")
eval_parser = subparsers.add_parser("eval")
# Predict (2D pose and/or 3D location from images)
predict_parser.add_argument('images', nargs='*', help='input images')
predict_parser.add_argument(
'--glob', help='glob expression for input images (for many images)')
predict_parser.add_argument('-o',
'--output-directory',
help='Output directory')
predict_parser.add_argument(
'--output_types',
nargs='+',
default=['json'],
help='what to output: json keypoints skeleton for Pifpaf'
'json bird front or multi for MonStereo')
predict_parser.add_argument('--no_save',
help='to show images',
action='store_true')
predict_parser.add_argument('--dpi',
help='image resolution',
type=int,
default=150)
predict_parser.add_argument(
'--long-edge',
default=None,
type=int,
help='rescale the long side of the image (aspect ratio maintained)')
predict_parser.add_argument(
'--white-overlay',
nargs='?',
default=False,
const=0.8,
type=float,
help='increase contrast to annotations by making image whiter')
predict_parser.add_argument('--font-size',
default=0,
type=int,
help='annotation font size')
predict_parser.add_argument('--monocolor-connections',
default=False,
action='store_true',
help='use a single color per instance')
predict_parser.add_argument('--instance-threshold',
type=float,
default=None,
help='threshold for entire instance')
predict_parser.add_argument('--seed-threshold',
type=float,
default=0.5,
help='threshold for single seed')
predict_parser.add_argument('--disable-cuda',
action='store_true',
help='disable CUDA')
predict_parser.add_argument(
'--focal',
help=
'focal length in mm for a sensor size of 7.2x5.4 mm. Default nuScenes sensor',
type=float,
default=5.7)
decoder.cli(parser)
logger.cli(parser)
network.Factory.cli(parser)
show.cli(parser)
visualizer.cli(parser)
predict_parser.add_argument('--mode',
help='keypoints, mono, stereo',
default='mono')
predict_parser.add_argument(
'--model', help='path of MonoLoco/MonStereo model to load')
predict_parser.add_argument(
'--net',
help='only to select older MonoLoco model, otherwise use --mode')
predict_parser.add_argument(
'--path_gt', help='path of json file with gt 3d localization')
predict_parser.add_argument('--z_max',
type=int,
help='maximum meters distance for predictions',
default=100)
predict_parser.add_argument('--n_dropout',
type=int,
help='Epistemic uncertainty evaluation',
default=0)
predict_parser.add_argument('--dropout',
type=float,
help='dropout parameter',
default=0.2)
predict_parser.add_argument(
'--show_all',
help='only predict ground-truth matches or all',
action='store_true')
# Social distancing and social interactions
predict_parser.add_argument('--social_distance',
help='social',
action='store_true')
predict_parser.add_argument('--threshold_prob',
type=float,
help='concordance for samples',
default=0.25)
predict_parser.add_argument('--threshold_dist',
type=float,
help='min distance of people',
default=2.5)
predict_parser.add_argument('--radii',
type=tuple,
help='o-space radii',
default=(0.3, 0.5, 1))
# Preprocess input data
prep_parser.add_argument('--dir_ann',
help='directory of annotations of 2d joints',
required=True)
prep_parser.add_argument('--mode', help='mono, stereo', default='mono')
prep_parser.add_argument(
'--dataset',
help=
'datasets to preprocess: nuscenes, nuscenes_teaser, nuscenes_mini, kitti',
default='kitti')
prep_parser.add_argument('--dir_nuscenes',
help='directory of nuscenes devkit',
default='data/nuscenes/')
prep_parser.add_argument('--iou_min',
help='minimum iou to match ground truth',
type=float,
default=0.3)
prep_parser.add_argument('--variance', help='new', action='store_true')
prep_parser.add_argument('--activity', help='new', action='store_true')
# Training
training_parser.add_argument('--joints',
help='Json file with input joints',
required=True)
training_parser.add_argument('--mode', help='mono, stereo', default='mono')
training_parser.add_argument('-e',
'--epochs',
type=int,
help='number of epochs to train for',
default=500)
training_parser.add_argument('--bs',
type=int,
default=512,
help='input batch size')
training_parser.add_argument('--monocular',
help='whether to train monoloco',
action='store_true')
training_parser.add_argument('--dropout',
type=float,
help='dropout. Default no dropout',
default=0.2)
training_parser.add_argument('--lr',
type=float,
help='learning rate',
default=0.001)
training_parser.add_argument('--sched_step',
type=float,
help='scheduler step time (epochs)',
default=30)
training_parser.add_argument('--sched_gamma',
type=float,
help='Scheduler multiplication every step',
default=0.98)
training_parser.add_argument('--hidden_size',
type=int,
help='Number of hidden units in the model',
default=1024)
training_parser.add_argument('--n_stage',
type=int,
help='Number of stages in the model',
default=3)
training_parser.add_argument('--hyp',
help='run hyperparameters tuning',
action='store_true')
training_parser.add_argument('--multiplier',
type=int,
help='Size of the grid of hyp search',
default=1)
training_parser.add_argument(
'--r_seed',
type=int,
help='specify the seed for training and hyp tuning',
default=1)
training_parser.add_argument('--print_loss',
help='print training and validation losses',
action='store_true')
training_parser.add_argument(
'--auto_tune_mtl',
help='whether to use uncertainty to autotune losses',
action='store_true')
training_parser.add_argument('--no_save',
help='to not save model and log file',
action='store_true')
# Evaluation
eval_parser.add_argument('--mode', help='mono, stereo', default='mono')
eval_parser.add_argument('--dataset',
help='datasets to evaluate, kitti or nuscenes',
default='kitti')
eval_parser.add_argument('--activity',
help='evaluate activities',
action='store_true')
eval_parser.add_argument('--geometric',
help='to evaluate geometric distance',
action='store_true')
eval_parser.add_argument('--generate',
help='create txt files for KITTI evaluation',
action='store_true')
eval_parser.add_argument(
'--dir_ann',
help='directory of annotations of 2d joints (for KITTI evaluation)')
eval_parser.add_argument('--model', help='path of MonoLoco model to load')
eval_parser.add_argument(
'--joints',
help='Json file with input joints to evaluate (for nuScenes evaluation)'
)
eval_parser.add_argument('--n_dropout',
type=int,
help='Epistemic uncertainty evaluation',
default=0)
eval_parser.add_argument('--dropout',
type=float,
help='dropout. Default no dropout',
default=0.2)
eval_parser.add_argument('--hidden_size',
type=int,
help='Number of hidden units in the model',
default=1024)
eval_parser.add_argument('--n_stage',
type=int,
help='Number of stages in the model',
default=3)
eval_parser.add_argument('--show',
help='whether to show statistic graphs',
action='store_true')
eval_parser.add_argument('--save',
help='whether to save statistic graphs',
action='store_true')
eval_parser.add_argument('--verbose',
help='verbosity of statistics',
action='store_true')
eval_parser.add_argument('--new', help='new', action='store_true')
eval_parser.add_argument('--variance',
help='evaluate keypoints variance',
action='store_true')
eval_parser.add_argument(
'--net',
help='Choose network: monoloco, monoloco_p, monoloco_pp, monstereo')
eval_parser.add_argument('--baselines',
help='whether to evaluate stereo baselines',
action='store_true')
eval_parser.add_argument(
'--generate_official',
help='whether to add empty txt files for official evaluation',
action='store_true')
args = parser.parse_args()
return args
def cli():
parser = argparse.ArgumentParser(
prog='python3 -m openpifpaf.predict',
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
nets.cli(parser)
decoder.cli(parser,
force_complete_pose=False,
instance_threshold=0.1,
seed_threshold=0.5)
parser.add_argument('images', nargs='*', help='input images')
parser.add_argument(
'--glob', help='glob expression for input images (for many images)')
parser.add_argument('-o',
'--output-directory',
help=('Output directory. When using this option, make '
'sure input images have distinct file names.'))
parser.add_argument('--show',
default=False,
action='store_true',
help='show image of output overlay')
parser.add_argument('--output-types',
nargs='+',
default=['skeleton', 'json'],
help='what to output: skeleton, keypoints, json')
parser.add_argument('--batch-size',
default=1,
type=int,
help='processing batch size')
parser.add_argument('--long-edge',
default=None,
type=int,
help='apply preprocessing to batch images')
parser.add_argument('--loader-workers',
default=None,
type=int,
help='number of workers for data loading')
parser.add_argument('--disable-cuda',
action='store_true',
help='disable CUDA')
parser.add_argument('--line-width',
default=6,
type=int,
help='line width for skeleton')
parser.add_argument('--figure-width',
default=10.0,
type=float,
help='figure width')
parser.add_argument('--dpi-factor',
default=1.0,
type=float,
help='increase dpi of output image by this factor')
parser.add_argument('--our-new-model',
default=0.0,
type=float,
help='uses our new model')
group = parser.add_argument_group('logging')
group.add_argument('-q',
'--quiet',
default=False,
action='store_true',
help='only show warning messages or above')
group.add_argument('--debug',
default=False,
action='store_true',
help='print debug messages')
args = parser.parse_args()
log_level = logging.INFO
if args.quiet:
log_level = logging.WARNING
if args.debug:
log_level = logging.DEBUG
logging.basicConfig()
logging.getLogger('openpifpaf').setLevel(log_level)
LOG.setLevel(log_level)
if args.loader_workers is None:
args.loader_workers = args.batch_size
# glob
if args.glob:
args.images += glob.glob(args.glob)
if not args.images:
raise Exception("no image files given")
# add args.device
args.device = torch.device('cpu')
args.pin_memory = False
if not args.disable_cuda and torch.cuda.is_available():
args.device = torch.device('cuda')
args.pin_memory = True
return args
def cli():
parser = argparse.ArgumentParser(
prog='python3 -m openpifpaf.predict',
usage='%(prog)s [options] images',
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
)
parser.add_argument('--version', action='version',
version='OpenPifPaf {version}'.format(version=__version__))
decoder.cli(parser)
logger.cli(parser)
network.Factory.cli(parser)
show.cli(parser)
visualizer.cli(parser)
parser.add_argument('images', nargs='*',
help='input images')
parser.add_argument('--glob',
help='glob expression for input images (for many images)')
parser.add_argument('-o', '--image-output', default=None, nargs='?', const=True,
help='Whether to output an image, '
'with the option to specify the output path or directory')
parser.add_argument('--json-output', default=None, nargs='?', const=True,
help='Whether to output a json file, '
'with the option to specify the output path or directory')
parser.add_argument('--batch-size', default=1, type=int,
help='processing batch size')
parser.add_argument('--long-edge', default=None, type=int,
help='rescale the long side of the image (aspect ratio maintained)')
parser.add_argument('--loader-workers', default=None, type=int,
help='number of workers for data loading')
parser.add_argument('--precise-rescaling', dest='fast_rescaling',
default=True, action='store_false',
help='use more exact image rescaling (requires scipy)')
parser.add_argument('--disable-cuda', action='store_true',
help='disable CUDA')
args = parser.parse_args()
logger.configure(args, LOG) # logger first
# add args.device
args.device = torch.device('cpu')
args.pin_memory = False
if not args.disable_cuda and torch.cuda.is_available():
args.device = torch.device('cuda')
args.pin_memory = True
LOG.info('neural network device: %s (CUDA available: %s, count: %d)',
args.device, torch.cuda.is_available(), torch.cuda.device_count())
decoder.configure(args)
network.Factory.configure(args)
show.configure(args)
visualizer.configure(args)
if args.loader_workers is None:
args.loader_workers = args.batch_size
# glob
if args.glob:
args.images += glob.glob(args.glob)
if not args.images:
raise Exception("no image files given")
return args
