File size: 13,856 Bytes
e437acb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 |
from __future__ import print_function, division
import torch
import matplotlib.pyplot as plt
import argparse, os
import numpy as np
from torch.utils.data import DataLoader
from torchvision import transforms
from models.CDCNs_u import Conv2d_cd, CDCN_u
from Load_OULUNPUcrop_train import Spoofing_train_g, SeparateBatchSampler, Normaliztion, ToTensor, \
RandomHorizontalFlip, Cutout, RandomErasing
from Load_OULUNPUcrop_valtest import Spoofing_valtest, Normaliztion_valtest, ToTensor_valtest
import torch.nn.functional as F
import torch.nn as nn
import torch.optim as optim
from utils import AvgrageMeter, performances
# Dataset root
train_image_dir = '/export2/home/wht/oulu_img_crop/train_file_flod/'
val_image_dir = '/export2/home/wht/oulu_img_crop/dev_file_flod/'
test_image_dir = '/export2/home/wht/oulu_img_crop/test_file_flod/'
train_map_dir = '/export2/home/wht/oulu_img_crop/train_depth_flod/'
val_map_dir = '/export2/home/wht/oulu_img_crop/dev_depth_flod/'
test_map_dir = '/export2/home/wht/oulu_img_crop/test_depth_flod/'
train_list = '/export2/home/wht/oulu_img_crop/protocols/Protocol_1/Train_g.txt'
val_list = '/export2/home/wht/oulu_img_crop/protocols/Protocol_1/Dev.txt'
test_list = '/export2/home/wht/oulu_img_crop/protocols/Protocol_1/Test.txt'
def contrast_depth_conv(input):
''' compute contrast depth in both of (out, label) '''
'''
input 32x32
output 8x32x32
'''
kernel_filter_list = [
[[1, 0, 0], [0, -1, 0], [0, 0, 0]], [[0, 1, 0], [0, -1, 0], [0, 0, 0]], [[0, 0, 1], [0, -1, 0], [0, 0, 0]],
[[0, 0, 0], [1, -1, 0], [0, 0, 0]], [[0, 0, 0], [0, -1, 1], [0, 0, 0]],
[[0, 0, 0], [0, -1, 0], [1, 0, 0]], [[0, 0, 0], [0, -1, 0], [0, 1, 0]], [[0, 0, 0], [0, -1, 0], [0, 0, 1]]
]
kernel_filter = np.array(kernel_filter_list, np.float32)
kernel_filter = torch.from_numpy(kernel_filter.astype(np.float)).float().cuda()
# weights (in_channel, out_channel, kernel, kernel)
kernel_filter = kernel_filter.unsqueeze(dim=1)
input = input.unsqueeze(dim=1).expand(input.shape[0], 8, input.shape[1], input.shape[2])
contrast_depth = F.conv2d(input, weight=kernel_filter, groups=8)
return contrast_depth
class Contrast_depth_loss(nn.Module):
def __init__(self):
super(Contrast_depth_loss, self).__init__()
return
def forward(self, out, label):
contrast_out = contrast_depth_conv(out)
contrast_label = contrast_depth_conv(label)
criterion_MSE = nn.MSELoss().cuda()
loss = criterion_MSE(contrast_out, contrast_label)
return loss
def train_test():
isExists = os.path.exists(args.log)
if not isExists:
os.makedirs(args.log)
log_file = open(args.log + '/' + args.log + '_log_P1.txt', 'a')
log_file.write('Oulu-NPU, P1:\n ')
log_file.flush()
print('train from scratch!\n')
log_file.write('train from scratch!\n')
log_file.write('lr:%.6f, lamda_kl:%.6f , batchsize:%d\n' % (args.lr, args.kl_lambda, args.batchsize))
log_file.flush()
model = CDCN_u(basic_conv=Conv2d_cd, theta=0.7)
# model = ResNet18_u()
model = model.cuda()
model = torch.nn.DataParallel(model)
lr = args.lr
optimizer = optim.Adam(model.parameters(), lr=lr, weight_decay=0.00005)
scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=args.step_size, gamma=args.gamma)
print(model)
criterion_absolute_loss = nn.MSELoss().cuda()
criterion_contrastive_loss = Contrast_depth_loss().cuda()
for epoch in range(args.epochs):
if (epoch + 1) % args.step_size == 0:
lr *= args.gamma
loss_absolute_real = AvgrageMeter()
loss_absolute_fake = AvgrageMeter()
loss_contra_real = AvgrageMeter()
loss_contra_fake = AvgrageMeter()
loss_kl_real = AvgrageMeter()
loss_kl_fake = AvgrageMeter()
###########################################
''' train '''
###########################################
model.train()
# load random 16-frame clip data every epoch
train_data = Spoofing_train_g(train_list, train_image_dir, train_map_dir,
transform=transforms.Compose(
[RandomErasing(), RandomHorizontalFlip(), ToTensor(), Cutout(),
Normaliztion()]))
train_real_idx, train_fake_idx = train_data.get_idx()
batch_sampler = SeparateBatchSampler(train_real_idx, train_fake_idx, batch_size=args.batchsize, ratio=args.ratio)
dataloader_train = DataLoader(train_data, num_workers=8, batch_sampler=batch_sampler)
for i, sample_batched in enumerate(dataloader_train):
# get the inputs
inputs, map_label, spoof_label = sample_batched['image_x'].cuda(), sample_batched['map_x'].cuda(), \
sample_batched['spoofing_label'].cuda()
optimizer.zero_grad()
# forward + backward + optimize
mu, logvar, map_x, x_concat, x_Block1, x_Block2, x_Block3, x_input = model(inputs)
mu_real = mu[:int(args.batchsize * args.ratio), :, :]
logvar_real = logvar[:int(args.batchsize * args.ratio), :, :]
map_x_real = map_x[:int(args.batchsize * args.ratio), :, :]
map_label_real = map_label[:int(args.batchsize * args.ratio), :, :]
absolute_loss_real = criterion_absolute_loss(map_x_real, map_label_real)
contrastive_loss_real = criterion_contrastive_loss(map_x_real, map_label_real)
kl_loss_real = -(1 + logvar_real - (mu_real - map_label_real).pow(2) - logvar_real.exp()) / 2
kl_loss_real = kl_loss_real.sum(dim=1).sum(dim=1).mean()
kl_loss_real = args.kl_lambda * kl_loss_real
mu_fake = mu[int(args.batchsize * args.ratio):, :, :]
logvar_fake = logvar[int(args.batchsize * args.ratio):, :, :]
map_x_fake = map_x[int(args.batchsize * args.ratio):, :, :]
map_label_fake = map_label[int(args.batchsize * args.ratio):, :, :]
absolute_loss_fake = 0.1 * criterion_absolute_loss(map_x_fake, map_label_fake)
contrastive_loss_fake = 0.1 * criterion_contrastive_loss(map_x_fake, map_label_fake)
kl_loss_fake = -(1 + logvar_fake - (mu_fake - map_label_fake).pow(2) - logvar_fake.exp()) / 2
kl_loss_fake = kl_loss_fake.sum(dim=1).sum(dim=1).mean()
kl_loss_fake = 0.1 * args.kl_lambda * kl_loss_fake
absolute_loss = absolute_loss_real + absolute_loss_fake
contrastive_loss = contrastive_loss_real + contrastive_loss_fake
kl_loss = kl_loss_real + kl_loss_fake
loss = absolute_loss + contrastive_loss + kl_loss
loss.backward()
optimizer.step()
n = inputs.size(0)
loss_absolute_real.update(absolute_loss_real.data, n)
loss_absolute_fake.update(absolute_loss_fake.data, n)
loss_contra_real.update(contrastive_loss_real.data, n)
loss_contra_fake.update(contrastive_loss_fake.data, n)
loss_kl_real.update(kl_loss_real.data, n)
loss_kl_fake.update(kl_loss_fake.data, n)
scheduler.step()
# whole epoch average
print(
'epoch:%d, Train: Absolute_loss: real=%.4f,fake=%.4f, '
'Contrastive_loss: real=%.4f,fake=%.4f, kl_loss: real=%.4f,fake=%.4f' % (
epoch + 1, loss_absolute_real.avg, loss_absolute_fake.avg, loss_contra_real.avg, loss_contra_fake.avg,
loss_kl_real.avg, loss_kl_fake.avg))
# validation/test
if epoch < 200:
epoch_test = 200
else:
epoch_test = 50
# epoch_test = 1
if epoch % epoch_test == epoch_test - 1:
model.eval()
with torch.no_grad():
###########################################
''' val '''
###########################################
# val for threshold
val_data = Spoofing_valtest(val_list, val_image_dir, val_map_dir,
transform=transforms.Compose([Normaliztion_valtest(), ToTensor_valtest()]))
dataloader_val = DataLoader(val_data, batch_size=1, shuffle=False, num_workers=4)
map_score_list = []
for i, sample_batched in enumerate(dataloader_val):
# get the inputs
inputs, spoof_label = sample_batched['image_x'].cuda(), sample_batched['spoofing_label'].cuda()
val_maps = sample_batched['val_map_x'].cuda() # binary map from PRNet
optimizer.zero_grad()
mu, logvar, map_x, x_concat, x_Block1, x_Block2, x_Block3, x_input = model(inputs.squeeze(0))
score_norm = mu.sum(dim=1).sum(dim=1) / val_maps.squeeze(0).sum(dim=1).sum(dim=1)
map_score = score_norm.mean()
map_score_list.append('{} {}\n'.format(map_score, spoof_label[0][0]))
map_score_val_filename = args.log + '/' + args.log + '_map_score_val.txt'
with open(map_score_val_filename, 'w') as file:
file.writelines(map_score_list)
###########################################
''' test '''
##########################################
# test for ACC
test_data = Spoofing_valtest(test_list, test_image_dir, test_map_dir,
transform=transforms.Compose([Normaliztion_valtest(), ToTensor_valtest()]))
dataloader_test = DataLoader(test_data, batch_size=1, shuffle=False, num_workers=4)
map_score_list = []
for i, sample_batched in enumerate(dataloader_test):
# get the inputs
inputs, spoof_label = sample_batched['image_x'].cuda(), sample_batched['spoofing_label'].cuda()
test_maps = sample_batched['val_map_x'].cuda()
optimizer.zero_grad()
mu, logvar, map_x, x_concat, x_Block1, x_Block2, x_Block3, x_input = model(inputs.squeeze(0))
score_norm = mu.sum(dim=1).sum(dim=1) / test_maps.squeeze(0).sum(dim=1).sum(dim=1)
map_score = score_norm.mean()
map_score_list.append('{} {}\n'.format(map_score, spoof_label[0][0]))
map_score_test_filename = args.log + '/' + args.log + '_map_score_test.txt'
with open(map_score_test_filename, 'w') as file:
file.writelines(map_score_list)
#############################################################
# performance measurement both val and test
#############################################################
val_threshold, test_threshold, val_ACC, val_ACER, test_ACC, test_APCER, test_BPCER, test_ACER, test_ACER_test_threshold = performances(
map_score_val_filename, map_score_test_filename)
print('epoch:%d, Val: val_threshold= %.4f, val_ACC= %.4f, val_ACER= %.4f' % (
epoch + 1, val_threshold, val_ACC, val_ACER))
log_file.write('\n epoch:%d, Val: val_threshold= %.4f, val_ACC= %.4f, val_ACER= %.4f \n' % (
epoch + 1, val_threshold, val_ACC, val_ACER))
print('epoch:%d, Test: ACC= %.4f, APCER= %.4f, BPCER= %.4f, ACER= %.4f' % (
epoch + 1, test_ACC, test_APCER, test_BPCER, test_ACER))
log_file.write('epoch:%d, Test: ACC= %.4f, APCER= %.4f, BPCER= %.4f, ACER= %.4f \n' % (
epoch + 1, test_ACC, test_APCER, test_BPCER, test_ACER))
log_file.flush()
if epoch % epoch_test == epoch_test - 1:
# save the model until the next improvement
torch.save(model.state_dict(), args.log + '/' + args.log + '_%d.pkl' % (epoch + 1))
print('Finished Training')
log_file.close()
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="save quality using landmarkpose model")
parser.add_argument('--gpus', type=str, default='0, 1, 2, 3', help='the gpu id used for predict')
parser.add_argument('--lr', type=float, default=0.0001, help='initial learning rate')
parser.add_argument('--batchsize', type=int, default=64, help='initial batchsize')
parser.add_argument('--step_size', type=int, default=500, help='how many epochs lr decays once') # 500
parser.add_argument('--gamma', type=float, default=0.5, help='gamma of optim.lr_scheduler.StepLR, decay of lr')
parser.add_argument('--kl_lambda', type=float, default=0.001, help='')
parser.add_argument('--ratio', type=float, default=0.75, help='real and fake in batchsize ')
parser.add_argument('--echo_batches', type=int, default=50, help='how many batches display once') # 50
parser.add_argument('--epochs', type=int, default=1600, help='total training epochs')
parser.add_argument('--log', type=str, default="CDCN_U_P1", help='log and save model name')
parser.add_argument('--finetune', action='store_true', default=False, help='whether finetune other models')
args = parser.parse_args()
train_test()
|