im_name=str(i) + '.jpg')
ckpt_time, post_time = getTime(ckpt_time)
runtime_profile['pn'].append(post_time)
# TQDM
im_names_desc.set_description(
'load time: {ld:.4f} | det time: {dt:.4f} | det NMS: {dn:.4f} | pose time: {pt:.4f} | post process: {pn:.4f}'
.format(ld=np.mean(runtime_profile['ld']),
dt=np.mean(runtime_profile['dt']),
dn=np.mean(runtime_profile['dn']),
pt=np.mean(runtime_profile['pt']),
pn=np.mean(runtime_profile['pn'])))
except KeyboardInterrupt:
break
print(' ')
print('===========================> Finish Model Running.')
if (args.save_img or args.save_video) and not args.vis_fast:
print(
'===========================> Rendering remaining images in the queue...'
)
print(
'===========================> If this step takes too long, you can enable the --vis_fast flag to use fast rendering (real-time).'
)
while (writer.running()):
pass
writer.stop()
final_result = writer.results()
write_json(final_result, args.outputpath)
def call_alphapose(input_dir, output_dir, format='open', batchSize=1):
if not os.path.exists(output_dir):
os.mkdir(output_dir)
for root, dirs, files in os.walk(input_dir):
im_names = files
print(files)
data_loader = ImageLoader(im_names,
batchSize=batchSize,
format='yolo',
dir_path=input_dir).start()
det_loader = DetectionLoader(data_loader, batchSize=batchSize).start()
det_processor = DetectionProcessor(det_loader).start()
# Load pose model
pose_dataset = Mscoco()
pose_model = InferenNet(4 * 1 + 1, pose_dataset)
pose_model.cuda()
pose_model.eval()
runtime_profile = {'dt': [], 'pt': [], 'pn': []}
# Init data writer
writer = DataWriter(False).start()
data_len = data_loader.length()
im_names_desc = tqdm(range(data_len))
for i in im_names_desc:
start_time = getTime()
with torch.no_grad():
(inps, orig_img, im_name, boxes, scores, pt1,
pt2) = det_processor.read()
if boxes is None or boxes.nelement() == 0:
writer.save(None, None, None, None, None, orig_img,
im_name.split('/')[-1])
continue
ckpt_time, det_time = getTime(start_time)
runtime_profile['dt'].append(det_time)
# Pose Estimation
datalen = inps.size(0)
leftover = 0
if (datalen) % batchSize:
leftover = 1
num_batches = datalen // batchSize + leftover
hm = []
for j in range(num_batches):
inps_j = inps[j * batchSize:min((j + 1) *
batchSize, datalen)].cuda()
hm_j = pose_model(inps_j)
hm.append(hm_j)
hm = torch.cat(hm)
ckpt_time, pose_time = getTime(ckpt_time)
runtime_profile['pt'].append(pose_time)
hm = hm.cpu()
writer.save(boxes, scores, hm, pt1, pt2, orig_img,
im_name.split('/')[-1])
ckpt_time, post_time = getTime(ckpt_time)
runtime_profile['pn'].append(post_time)
while (writer.running()):
pass
writer.stop()
final_result = writer.results()
write_json(final_result, output_dir, _format=format)
correct_json_save(output_dir)
print('Over')
for i in im_names_desc:
with torch.no_grad():
(inps, orig_img, im_name, boxes, scores, pt1, pt2) = det_processor.Q[i]
if orig_img is None:
break
if boxes is None or boxes.nelement() == 0:
writer.save(None, None, None, None, None, orig_img, im_name)
continue
# Pose Estimation
datalen = inps.size(0)
leftover = 0
if (datalen) % batchSize:
leftover = 1
num_batches = datalen // batchSize + leftover
hm = []
for j in range(num_batches):
if opt.device == 'GPU':
inps_j = inps[j * batchSize:min((j + 1) * batchSize, datalen)].cuda()
else:
inps_j = inps[j * batchSize:min((j + 1) * batchSize, datalen)].cpu()
hm_j = pose_model(inps_j)
hm.append(hm_j)
hm = torch.cat(hm)
hm = hm.cpu().data
writer.save(boxes, scores, hm, pt1, pt2, orig_img, im_name)
writer.start()
final_result = writer.results()
write_json(final_result, os.path.join(args.outputpath, ntpath.basename(args.video).split('.')[0] + '.json'))
hm = hm.cpu().data
writer.save(boxes, scores, hm, pt1, pt2, orig_img, im_name, CAR)
ckpt_time, post_time = getTime(ckpt_time)
runtime_profile['pn'].append(post_time)
# if args.profile:
# im_names_desc.set_description(
# 'det time: {dt:.3f} | pose time: {pt:.2f} | post processing: {pn:.4f}'.format(
# dt=np.mean(runtime_profile['dt']), pt=np.mean(runtime_profile['pt']),
# pn=np.mean(runtime_profile['pn']))
# )
print('===========================> Finish Model Running.')
if (args.save_img or args.save_video) and not args.vis_fast:
print(
'===========================> Rendering remaining images in the queue...'
)
print(
'===========================> If this step takes too long, you can enable the --vis_fast flag to use fast rendering (real-time).'
)
while (writer.running()):
pass
writer.stop()
writer.write_enery()
final_result = writer.results()
# writer.join()
# time.sleep(3)
write_json(videofile, final_result, args.outputpath)
def test():
inputpath = args.inputpath
inputlist = args.inputlist
mode = args.mode
#if not os.path.exists(args.outputpath):
#os.mkdir(args.outputpath)
#if len(inputlist):
#im_names = open(inputlist, 'r').readlines()
#elif len(inputpath) and inputpath != '/':
for root, dirs, files in os.walk(inputpath):
im_names = files
#else:
#raise IOError('Error: must contain either --indir/--list')
im_names = sorted(im_names, key=lambda x: int(os.path.splitext(x)[0]))
print(im_names)
# Load input images
data_loader = ImageLoader(im_names, batchSize=1, format='yolo').start()
# Load detection loader
print('Loading YOLO model..')
sys.stdout.flush()
det_loader = DetectionLoader(data_loader, batchSize=1).start()
det_processor = DetectionProcessor(det_loader).start()
runtime_profile = {'dt': [], 'pt': [], 'pn': []}
# Init data writer
writer = DataWriter(args.save_video).start()
data_len = data_loader.length()
im_names_desc = tqdm(range(data_len))
batchSize = args.posebatch
for i in im_names_desc:
start_time = getTime()
with torch.no_grad():
(inps, orig_img, im_name, boxes, scores, pt1,
pt2) = det_processor.read()
if boxes is None or boxes.nelement() == 0:
writer.save(None, None, None, None, None, orig_img,
im_name.split('/')[-1])
continue
ckpt_time, det_time = getTime(start_time)
runtime_profile['dt'].append(det_time)
# Pose Estimation
datalen = inps.size(0)
leftover = 0
if (datalen) % batchSize:
leftover = 1
num_batches = datalen // batchSize + leftover
hm = []
for j in range(num_batches):
inps_j = inps[j * batchSize:min((j + 1) *
batchSize, datalen)].cuda()
hm_j = pose_model(inps_j)
hm.append(hm_j)
hm = torch.cat(hm)
ckpt_time, pose_time = getTime(ckpt_time)
runtime_profile['pt'].append(pose_time)
hm = hm.cpu()
writer.save(boxes, scores, hm, pt1, pt2, orig_img,
im_name.split('/')[-1])
ckpt_time, post_time = getTime(ckpt_time)
runtime_profile['pn'].append(post_time)
if args.profile:
# TQDM
im_names_desc.set_description(
'det time: {dt:.3f} | pose time: {pt:.2f} | post processing: {pn:.4f}'
.format(dt=np.mean(runtime_profile['dt']),
pt=np.mean(runtime_profile['pt']),
pn=np.mean(runtime_profile['pn'])))
print('===========================> Finish Model Running.')
if (args.save_img or args.save_video) and not args.vis_fast:
print(
'===========================> Rendering remaining images in the queue...'
)
print(
'===========================> If this step takes too long, you can enable the --vis_fast flag to use fast rendering (real-time).'
)
while (writer.running()):
pass
writer.stop()
final_result = writer.results()
write_json(final_result, args.outputpath)
return final_result
hm.append(hm_j)
hm = torch.cat(hm)
ckpt_time, pose_time = getTime(ckpt_time)
runtime_profile['pt'].append(pose_time)
hm = hm.cpu()
writer.save(boxes, scores, hm, pt1, pt2, orig_img, im_name.split('/')[-1])
ckpt_time, post_time = getTime(ckpt_time)
runtime_profile['pn'].append(post_time)
if args.profile:
# TQDM
im_names_desc.set_description(
'det time: {dt:.3f} | pose time: {pt:.2f} | post processing: {pn:.4f}'.format(
dt=np.mean(runtime_profile['dt']), pt=np.mean(runtime_profile['pt']), pn=np.mean(runtime_profile['pn']))
)
print('===========================> Finish Model Running.')
if (args.save_img or args.save_video) and not args.vis_fast:
print('===========================> Rendering remaining images in the queue...')
print('===========================> If this step takes too long, you can enable the --vis_fast flag to use fast rendering (real-time).')
while(writer.running()):
pass
writer.stop()
final_result = writer.results()
dest_path = os.path.join(args.outputpath,input_file_name)
if not os.path.exists(dest_path):
os.mkdir(dest_path)
write_json(final_result, dest_path)
def handle_video(video_file):
# =========== common ===============
args.video = video_file
base_name = os.path.basename(args.video)
video_name = base_name[:base_name.rfind('.')]
# =========== end common ===============
# =========== image ===============
# img_path = f'outputs/alpha_pose_{video_name}/split_image/'
# args.inputpath = img_path
# args.outputpath = f'outputs/alpha_pose_{video_name}'
# if os.path.exists(args.outputpath):
# shutil.rmtree(f'{args.outputpath}/vis', ignore_errors=True)
# else:
# os.mkdir(args.outputpath)
# # if not len(video_file):
# # raise IOError('Error: must contain --video')
# if len(img_path) and img_path != '/':
# for root, dirs, files in os.walk(img_path):
# im_names = sorted([f for f in files if 'png' in f or 'jpg' in f])
# else:
# raise IOError('Error: must contain either --indir/--list')
# # Load input images
# data_loader = ImageLoader(im_names, batchSize=args.detbatch, format='yolo').start()
# print(f'Totally {data_loader.datalen} images')
# =========== end image ===============
# =========== video ===============
args.outputpath = f'outputs/alpha_pose_{video_name}'
if os.path.exists(args.outputpath):
shutil.rmtree(f'{args.outputpath}/vis', ignore_errors=True)
else:
os.mkdir(args.outputpath)
videofile = args.video
mode = args.mode
if not len(videofile):
raise IOError('Error: must contain --video')
# Load input video
data_loader = VideoLoader(videofile, batchSize=args.detbatch).start()
(fourcc, fps, frameSize) = data_loader.videoinfo()
print('the video is {} f/s'.format(fps))
# =========== end video ===============
# Load detection loader
print('Loading YOLO model..')
sys.stdout.flush()
det_loader = DetectionLoader(data_loader, batchSize=args.detbatch).start()
# start a thread to read frames from the file video stream
det_processor = DetectionProcessor(det_loader).start()
# Load pose model
pose_dataset = Mscoco()
if args.fast_inference:
pose_model = InferenNet_fast(4 * 1 + 1, pose_dataset)
else:
pose_model = InferenNet(4 * 1 + 1, pose_dataset)
pose_model #.cuda()
pose_model.eval()
runtime_profile = {'dt': [], 'pt': [], 'pn': []}
# Data writer
save_path = os.path.join(
args.outputpath,
'AlphaPose_' + ntpath.basename(video_file).split('.')[0] + '.avi')
# writer = DataWriter(args.save_video, save_path, cv2.VideoWriter_fourcc(*'XVID'), fps, frameSize).start()
writer = DataWriter(args.save_video).start()
print('Start pose estimation...')
im_names_desc = tqdm(range(data_loader.length()))
batchSize = args.posebatch
for i in im_names_desc:
start_time = getTime()
with torch.no_grad():
(inps, orig_img, im_name, boxes, scores, pt1,
pt2) = det_processor.read()
if orig_img is None:
print(f'{i}-th image read None: handle_video')
break
if boxes is None or boxes.nelement() == 0:
writer.save(None, None, None, None, None, orig_img,
im_name.split('/')[-1])
continue
ckpt_time, det_time = getTime(start_time)
runtime_profile['dt'].append(det_time)
# Pose Estimation
datalen = inps.size(0)
leftover = 0
if datalen % batchSize:
leftover = 1
num_batches = datalen // batchSize + leftover
hm = []
for j in range(num_batches):
inps_j = inps[j * batchSize:min((j + 1) *
batchSize, datalen)] #.cuda()
hm_j = pose_model(inps_j)
hm.append(hm_j)
hm = torch.cat(hm)
ckpt_time, pose_time = getTime(ckpt_time)
runtime_profile['pt'].append(pose_time)
hm = hm.cpu().data
writer.save(boxes, scores, hm, pt1, pt2, orig_img,
im_name.split('/')[-1])
ckpt_time, post_time = getTime(ckpt_time)
runtime_profile['pn'].append(post_time)
if args.profile:
# TQDM
im_names_desc.set_description(
'det time: {dt:.4f} | pose time: {pt:.4f} | post processing: {pn:.4f}'
.format(dt=np.mean(runtime_profile['dt']),
pt=np.mean(runtime_profile['pt']),
pn=np.mean(runtime_profile['pn'])))
if (args.save_img or args.save_video) and not args.vis_fast:
print(
'===========================> Rendering remaining images in the queue...'
)
print(
'===========================> If this step takes too long, you can enable the --vis_fast flag to use fast rendering (real-time).'
)
while writer.running():
pass
writer.stop()
final_result = writer.results()
write_json(final_result, args.outputpath)
return final_result, video_name
try:
img, poses_T_1[k], pose = get_pose(k, Cap_T, flage=True)
align_pose_T = align_torso(poses_T_1[k])
poses_T_2[k] = change_dim(align_pose_T)
id_ = pose['imgname']
pose_id[k] = int(id_)
final_result.append(pose)
except:
print("number==>",k)
poses_T_1[k]=0
poses_T_2[k]=0
pose_id[k]=0
# continue
write_json(final_result, os.path.join(args.outputpath, name_T))
print("Video process completed ...")
elif not os.path.isfile(os.path.join(os.path.join(args.outputpath, name_T), "action_trainer_results.json")):
print("Warning: Your video is being processed for the first time and might take a few minutes ...")
final_result =[]
# k=0
for i in tqdm(range(total_len)):
try:
img, poses_T_1[i], pose = get_pose(i, Cap_T, flage=True)
align_pose_T = align_torso(poses_T_1[i])
poses_T_2[i] = change_dim(align_pose_T)
k = pose['imgname']
pose_id[i] = int(k)
final_result.append(pose)
hm = pose_model(inps)
ckpt_time, pose_time = getTime(ckpt_time)
runtime_profile['pt'].append(pose_time)
writer.save(boxes, scores, hm, pt1, pt2, orig_img,
im_name.split('/')[-1])
print("writer:", writer.len())
ckpt_time, post_time = getTime(ckpt_time)
runtime_profile['pn'].append(post_time)
# TQDM
im_names_desc.set_description(
'det time: {dt:.3f} | pose time: {pt:.2f} | post processing: {pn:.4f}'
.format(dt=np.mean(runtime_profile['dt']),
pt=np.mean(runtime_profile['pt']),
pn=np.mean(runtime_profile['pn'])))
print('===========================> Finish Model Running.')
if (args.save_img or args.save_video) and not args.vis_fast:
print(
'===========================> Rendering remaining images in the queue...'
)
print(
'===========================> If this step takes too long, you can enable the --vis_fast flag to use fast rendering (real-time).'
)
while (writer.running()):
pass
writer.stop()
final_result = writer.results()
write_json(final_result, args.outputpath, 'alphapose-results.json')
def main(file_name):
# videofile = args.video
videofile = file_name
mode = args.mode
if not os.path.exists(args.outputpath):
os.mkdir(args.outputpath)
if not len(videofile):
raise IOError('Error: must contain --video')
# Load input video
data_loader = VideoLoader(videofile, batchSize=args.detbatch).start()
(fourcc, fps, frameSize) = data_loader.videoinfo()
# Load detection loader
print('Loading YOLO model..')
sys.stdout.flush()
det_loader = DetectionLoader(data_loader, batchSize=args.detbatch).start()
det_processor = DetectionProcessor(det_loader).start()
# Load pose model
pose_dataset = Mscoco()
if args.fast_inference:
pose_model = InferenNet_fast(4 * 1 + 1, pose_dataset)
else:
pose_model = InferenNet(4 * 1 + 1, pose_dataset)
pose_model.cuda()
pose_model.eval()
runtime_profile = {'dt': [], 'pt': [], 'pn': []}
# Data writer
save_path = os.path.join(
args.outputpath,
'AlphaPose_' + ntpath.basename(videofile).split('.')[0] + '.avi')
writer = DataWriter(args.save_video, save_path,
cv2.VideoWriter_fourcc(*'XVID'), fps,
frameSize).start()
im_names_desc = tqdm(range(data_loader.length()))
batchSize = args.posebatch
for i in im_names_desc:
start_time = getTime()
with torch.no_grad():
(inps, orig_img, im_name, boxes, scores, pt1,
pt2) = det_processor.read()
if orig_img is None:
break
if boxes is None or boxes.nelement() == 0:
writer.save(None, None, None, None, None, orig_img,
im_name.split('/')[-1])
continue
ckpt_time, det_time = getTime(start_time)
runtime_profile['dt'].append(det_time)
# Pose Estimation
datalen = inps.size(0)
leftover = 0
if (datalen) % batchSize:
leftover = 1
num_batches = datalen // batchSize + leftover
hm = []
for j in range(num_batches):
inps_j = inps[j * batchSize:min((j + 1) *
batchSize, datalen)].cuda()
hm_j = pose_model(inps_j)
hm.append(hm_j)
hm = torch.cat(hm)
ckpt_time, pose_time = getTime(ckpt_time)
runtime_profile['pt'].append(pose_time)
hm = hm.cpu().data
import ipdb
ipdb.set_trace()
writer.save(boxes, scores, hm, pt1, pt2, orig_img,
im_name.split('/')[-1])
ckpt_time, post_time = getTime(ckpt_time)
runtime_profile['pn'].append(post_time)
if args.profile:
# TQDM
im_names_desc.set_description(
'det time: {dt:.4f} | pose time: {pt:.4f} | post processing: {pn:.4f}'
.format(dt=np.mean(runtime_profile['dt']),
pt=np.mean(runtime_profile['pt']),
pn=np.mean(runtime_profile['pn'])))
print('===========================> Finish Model Running.')
if (args.save_img or args.save_video) and not args.vis_fast:
print(
'===========================> Rendering remaining images in the queue...'
)
print(
'===========================> If this step takes too long, you can enable the --vis_fast flag to use fast rendering (real-time).'
)
while (writer.running()):
pass
writer.stop()
final_result = writer.results()
write_json(final_result, args.outputpath)
