def main(args):
args.use_smplx = True
device = torch.device(
'cuda') if torch.cuda.is_available() else torch.device('cpu')
assert torch.cuda.is_available(), "Current version only supports GPU"
#Set Bbox detector
bbox_detector = HandBboxDetector(args.view_type, device)
# Set Mocap regressor
hand_mocap = HandMocap(args.checkpoint_hand, args.smpl_dir, device=device)
# Set Visualizer
if args.renderer_type in ['pytorch3d', 'opendr']:
from renderer.screen_free_visualizer import Visualizer
else:
from renderer.visualizer import Visualizer
visualizer = Visualizer(args.renderer_type)
args.out_dir = os.path.join(
"/output/annotations/",
args.input_path.replace("/data/sessions_processed/final_dataset/",
"").split(".")[0], f"hand_bboxes_0.2")
print(args.out_dir)
assert os.path.exists(args.out_dir)
# run
run_hand_mocap(args, bbox_detector, hand_mocap, visualizer)
def main():
args = DemoOptions().parse()
args.use_smplx = True
device = torch.device(
'cuda') if torch.cuda.is_available() else torch.device('cpu')
assert torch.cuda.is_available(), "Current version only supports GPU"
hand_bbox_detector = HandBboxDetector('third_view', device)
#Set Mocap regressor
body_mocap = BodyMocap(args.checkpoint_body_smplx,
args.smpl_dir,
device=device,
use_smplx=True)
hand_mocap = HandMocap(args.checkpoint_hand, args.smpl_dir, device=device)
# Set Visualizer
if args.renderer_type in ['pytorch3d', 'opendr']:
from renderer.screen_free_visualizer import Visualizer
else:
from renderer.visualizer import Visualizer
visualizer = Visualizer(args.renderer_type)
run_frank_mocap(args, hand_bbox_detector, body_mocap, hand_mocap,
visualizer)
def __init__(self):
self.args = DemoOptions().parse()
self.args.use_smplx = True
self.args.save_pred_pkl = True
self.device = torch.device(
'cuda') if torch.cuda.is_available() else torch.device('cpu')
self.bbox_detector = HandBboxDetector(self.args.view_type, self.device)
self.hand_mocap = HandMocap(self.args.checkpoint_hand,
self.args.smpl_dir,
device=self.device)
self.visualizer = Visualizer(self.args.renderer_type)
def main():
args = DemoOptions().parse()
args.use_smplx = True
device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
assert torch.cuda.is_available(), "Current version only supports GPU"
hand_bbox_detector = HandBboxDetector('third_view', device)
#Set Mocap regressor
body_mocap = BodyMocap(args.checkpoint_body_smplx, args.smpl_dir, device = device, use_smplx= True)
hand_mocap = HandMocap(args.checkpoint_hand, args.smpl_dir, device = device)
run_frank_mocap(args, hand_bbox_detector, body_mocap, hand_mocap)
def run_from_list(args):
assert os.path.exists(args.list)
with open(args.list, "r") as f:
lines = f.readlines()
args.save_frame = False
args.no_video_out = True
args.use_smplx = True
device = torch.device(
'cuda') if torch.cuda.is_available() else torch.device('cpu')
assert torch.cuda.is_available(), "Current version only supports GPU"
#Set Bbox detector
bbox_detector = HandBboxDetector(args.view_type, device)
# Set Mocap regressor
hand_mocap = HandMocap(args.checkpoint_hand, args.smpl_dir, device=device)
# Set Visualizer
if args.renderer_type in ['pytorch3d', 'opendr']:
from renderer.screen_free_visualizer import Visualizer
else:
from renderer.visualizer import Visualizer
visualizer = Visualizer(args.renderer_type)
for i in range(1, len(lines)):
t0 = time.time()
path = lines[i].replace("\n", "")
session, task = (path.split(".")[0]).split("/")[-2:]
args.input_path = path
args.out_dir = f"/output/annotations/{session}/{task}/hand_bboxes_0.2/"
if not args.replace and os.path.exists(
os.path.join(args.out_dir, MARKER)):
print(
f"[{i}/{len(lines)-1}] Landmarks already extracted for this video -> '{args.out_dir}'"
)
continue
print(
f"[{i}/{len(lines)-1}] Extracting hand landmarks of '{session} - {task}'"
)
# run
run_hand_mocap(args, bbox_detector, hand_mocap, visualizer)
print(f"[FINISHED] Time: {time.time() - t0} s")
class HandExtractor:
def __init__(self, hand_checkpoint, smpl_folder):
self.hand_extractor = HandMocap(hand_checkpoint, smpl_folder)
def hands_from_df(self, img, hoa_df, resize_factor=1):
hand_df = hoa_df[hoa_df.det_type == "hand"]
hand_boxes = {}
# Keep first right and left hand found in data frame
for hand_idx, row in hand_df.iterrows():
box_ltrb = boxutils.dfbox_to_norm(row, resize_factor=resize_factor)
box_ltwh = [
box_ltrb[0],
box_ltrb[1],
box_ltrb[2] - box_ltrb[0],
box_ltrb[3] - box_ltrb[1],
]
box_side = row.side
hand_key = f"{box_side}_hand"
if hand_key not in hand_boxes:
hand_boxes[hand_key] = np.array(box_ltwh)
for side in ["right", "left"]:
hand_key = f"{side}_hand"
if hand_key not in hand_boxes:
hand_boxes[hand_key] = None
_, pred_hands = self.hand_extractor.regress(img, [hand_boxes],
add_margin=True)
# Retrieve first results
ego_hands = pred_hands[0]
pred_hands = {}
for side in ["right", "left"]:
hand_key = f"{side}_hand"
if hand_key in ego_hands and (ego_hands[hand_key] is not None):
pred_hand = ego_hands[hand_key]
pred_hands[side] = {
"verts": pred_hand["pred_vertices_img"],
"faces": pred_hand["faces"],
"hand_pose": pred_hand["pred_hand_pose"],
}
return pred_hands
def __init__(self, hand_checkpoint, smpl_folder):
self.hand_extractor = HandMocap(hand_checkpoint, smpl_folder)