parser.add_argument(
"--coco",
help="Do test on COCOPersons val set",
action="store_true",
)
parser.add_argument(
"--benchmark",
help="Do fps test without warpaffine",
action="store_true",
)
parser.add_argument(
"--OCHuman",
help="Do test on OCHuman val&test set",
action="store_true",
)
args = parser.parse_args()
jt.flags.use_cuda = 1
print('===========> loading model <===========')
model = Pose2Seg()
model.init(args.weights)
model.benchmark = args.benchmark
print('===========> testing <===========')
if args.coco:
test(model, dataset='cocoVal', benchmark=args.benchmark)
if args.OCHuman:
test(model, dataset='OCHumanVal', benchmark=args.benchmark)
test(model, dataset='OCHumanTest', benchmark=args.benchmark)
cocoEval.evaluate()
cocoEval.accumulate()
cocoEval.summarize()
return cocoEval
cocoEval = do_eval_coco(imgIds, datainfos.COCO, results_segm, 'segm')
logger('[POSE2SEG] AP|.5|.75| S| M| L| AR|.5|.75| S| M| L|')
_str = '[segm_score] %s '%dataset
for value in cocoEval.stats.tolist():
_str += '%.3f '%value
logger(_str)
if __name__=='__main__':
model_path = './pose2seg_release.pkl'
print('===========> loading model <===========')
model = Pose2Seg().cuda()
model.init(model_path)
model.eval()
ImageRoot = './data/images'
AnnoFile = './data/person_keypoints_val2017_pose2seg.json'
datainfos = CocoDatasetInfo(ImageRoot, AnnoFile, onlyperson=True, loadimg=True)
model.eval()
results_segm = []
imgIds = []
for i in tqdm(range(len(datainfos))):
rawdata = datainfos[i]
g1 = np.float32(rawdata['gt_keypoints'])
g2 = np.float32(rawdata['gt_keypoints']).transpose(0, 2, 1)
cur_canvas = canvas.copy()
cv2.fillConvexPoly(cur_canvas, polygon, colors[idx % len(colors)])
canvas = cv2.addWeighted(canvas, 0.4, cur_canvas, 0.6, 0)
return canvas
# Namespace(OCHuman=True, coco=True, weights='last.pkl')
cfg = AttrDict()
cfg.OCHuman = True
cfg.coco = True
cfg.weights = '/nethome/hkwon64/Research/imuTube/repos_v2/human_parsing/Pose2Seg/log/pose2seg_release.pkl'
device = torch.device('cuda', 1)
model = Pose2Seg(device).to(device)
model.init(cfg.weights)
model.eval()
dir_save = '/nethome/hkwon64/Research/imuTube/repos_v2/human_parsing/Pose2Seg/demo'
file_im = dir_save + '/demo.jpg'
file_ap = dir_save + '/alphapose-results.json'
ap_result = json.load(open(file_ap, 'r'))
gt_bbox = []
gt_kpts = []
for result in ap_result:
x1, y1, w, h = result['box']
bbox = int(x1), int(y1), int(x1 + w), int(y1 + h)
gt_bbox.append(bbox)