def load_cosface_model():
model_path = 'CosFace_pytorch/cosface_model.pth'
model = sphere().to(device)
model.load_state_dict(torch.load(model_path, map_location=device))
model.eval()
return model
def main():
# --------------------------------------model----------------------------------------
if args.network is 'sphere20':
model = net.sphere(type=20, is_gray=args.is_gray)
model_eval = net.sphere(type=20, is_gray=args.is_gray)
elif args.network is 'sphere64':
model = net.sphere(type=64, is_gray=args.is_gray)
model_eval = net.sphere(type=64, is_gray=args.is_gray)
elif args.network is 'LResNet50E_IR':
model = net.LResNet50E_IR(is_gray=args.is_gray)
model_eval = net.LResNet50E_IR(is_gray=args.is_gray)
else:
raise ValueError("NOT SUPPORT NETWORK! ")
model = torch.nn.DataParallel(model).to(device)
model_eval = model_eval.to(device)
print(model)
if not os.path.exists(args.save_path):
os.makedirs(args.save_path)
model.module.save(args.save_path + 'CosFace_0_checkpoint.pth')
# 512 is dimension of feature
classifier = {
'MCP': layer.MarginCosineProduct(512, args.num_class).to(device),
'AL': layer.AngleLinear(512, args.num_class).to(device),
'L': torch.nn.Linear(512, args.num_class, bias=False).to(device)
}[args.classifier_type]
# ------------------------------------load image---------------------------------------
if args.is_gray:
train_transform = transforms.Compose([
transforms.Grayscale(),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(), # range [0, 255] -> [0.0,1.0]
transforms.Normalize(mean=(0.5, ), std=(0.5, ))
]) # gray
else:
train_transform = transforms.Compose([
transforms.RandomHorizontalFlip(),
transforms.ToTensor(), # range [0, 255] -> [0.0,1.0]
transforms.Normalize(mean=(0.5, 0.5, 0.5),
std=(0.5, 0.5,
0.5)) # range [0.0, 1.0] -> [-1.0,1.0]
])
train_loader = torch.utils.data.DataLoader(ImageList(
root=args.root_path,
fileList=args.train_list,
transform=train_transform),
batch_size=args.batch_size,
shuffle=True,
num_workers=args.workers,
pin_memory=True,
drop_last=True)
print('length of train Database: ' + str(len(train_loader.dataset)))
print('Number of Identities: ' + str(args.num_class))
# --------------------------------loss function and optimizer-----------------------------
criterion = torch.nn.CrossEntropyLoss().to(device)
optimizer = torch.optim.SGD([{
'params': model.parameters()
}, {
'params': classifier.parameters()
}],
lr=args.lr,
momentum=args.momentum,
weight_decay=args.weight_decay)
# ----------------------------------------train----------------------------------------
# lfw_eval.eval(args.save_path + 'CosFace_0_checkpoint.pth')
for epoch in range(1, args.epochs + 1):
train(train_loader, model, classifier, criterion, optimizer, epoch)
model.module.save(args.save_path + 'CosFace_' + str(epoch) +
'_checkpoint.pth')
lfw_eval.eval(
model_eval,
args.save_path + 'CosFace_' + str(epoch) + '_checkpoint.pth',
args.is_gray)
print('Finished Training')
return ft_dict
def make_cosine_similarity_file(target_img, all_img_path, model, model_path):
'''
file_dir = 'experiments'
'''
img_list = os.listdir(all_img_path)
ft_dict = {}
txt_name = '_'.join(all_img_path.split('/'))
fo = open('experiments/{}.txt'.format(txt_name), 'w')
foo = open('experiments/min_max_{}.txt'.format(txt_name), 'w')
ft_dict = extract_all_img_ft(model, model_path, all_img_path, img_list)
make_cosine_similarity_matrix(ft_dict, img_list, fo, foo)
if __name__ == '__main__':
model = net.sphere().to('cuda')
model_path = 'checkpoint/original.pth'
img_path = 'lfw-112X96/Josh_Kronfeld/Josh_Kronfeld_0001.jpg'
CASIA_dataset = '/data1/aipd_tuijian/charlesliu/dataset/CASIA-WebFace'
lfw_dataset = '/data1/aipd_tuijian/charlesliu/dataset/lfw-112X96'
img_dir_name_india_women = ['6432391']
img_dir_name_men = ['6418193']
img_dir_name_india_women_glasses = ['027.jpg', '054.jpg']
img_dir_name_india_women_make_up = ['043.jpg', '037.jpg', '028.jpg', '016.jpg', '007.jpg']
img_dir_name_india_women_occlusion = ['023.jpg', '017.jpg']
img_1 = os.path.join(lfw_dataset, 'Abel_Pacheco', 'Abel_Pacheco_0001.jpg')
img_2 = os.path.join(lfw_dataset, 'Abel_Pacheco', 'Abel_Pacheco_0004.jpg')
f1 = extract_one_img_ft(model, model_path, img_1)
f2 = extract_one_img_ft(model, model_path, img_2)
distance = f1.dot(f2) / (f1.norm() * f2.norm() + 1e-5)
print distance
img1 = Image.open(f).convert('RGB')
with open(root + name2, 'rb') as f:
img2 = Image.open(f).convert('RGB')
f1 = extractDeepFeature(img1, model, is_gray)
f2 = extractDeepFeature(img2, model, is_gray)
distance = f1.dot(f2) / (f1.norm() * f2.norm() + 1e-5)
predicts.append('{}\t{}\t{}\t{}\n'.format(name1, name2, distance,
sameflag))
accuracy = []
thd = []
folds = KFold(n=6000, n_folds=10)
thresholds = np.arange(-1.0, 1.0, 0.005)
predicts = np.array(map(lambda line: line.strip('\n').split(), predicts))
for idx, (train, test) in enumerate(folds):
best_thresh = find_best_threshold(thresholds, predicts[train])
accuracy.append(eval_acc(best_thresh, predicts[test]))
thd.append(best_thresh)
print('LFWACC={:.4f} std={:.4f} thd={:.4f}'.format(np.mean(accuracy),
np.std(accuracy),
np.mean(thd)))
return np.mean(accuracy), predicts
if __name__ == '__main__':
_, result = eval(net.sphere().to('cuda'),
model_path='checkpoint/CosFace_24_checkpoint.pth')
np.savetxt("result.txt", result, '%s')
else:
raise ValueError("WRONG LINE IN 'pairs.txt! ")
with open(root + name1, 'rb') as f:
img1 = Image.open(f).convert('RGB')
with open(root + name2, 'rb') as f:
img2 = Image.open(f).convert('RGB')
f1 = extractDeepFeature(img1, model, is_gray)
f2 = extractDeepFeature(img2, model, is_gray)
distance = f1.dot(f2) / (f1.norm() * f2.norm() + 1e-5)
predicts.append('{}\t{}\t{}\t{}\n'.format(name1, name2, distance, sameflag))
accuracy = []
thd = []
folds = KFold(n=6000, n_folds=10)
thresholds = np.arange(-1.0, 1.0, 0.005)
predicts = np.array(map(lambda line: line.strip('\n').split(), predicts))
for idx, (train, test) in enumerate(folds):
best_thresh = find_best_threshold(thresholds, predicts[train])
accuracy.append(eval_acc(best_thresh, predicts[test]))
thd.append(best_thresh)
print('LFWACC={:.4f} std={:.4f} thd={:.4f}'.format(np.mean(accuracy), np.std(accuracy), np.mean(thd)))
return np.mean(accuracy), predicts
if __name__ == '__main__':
_, result = eval(net.sphere().to('cuda'), model_path='checkpoint/CosFace_24_checkpoint.pth')
np.savetxt("result.txt", result, '%s')