def main(args):
image_paths = read_list(args.image_list)
for path in image_paths:
# landmarks_file should have the same prefix as image_file
landmarks_file = path[:-3] + 'txt'
im = Image.open(path)
width, height = im.size
landmarks = load_landmarks(landmarks_file)
landmarks[:,1] = height - landmarks[:,1]
# select contour points
#contour_points = get_contour_side(landmarks)
# generate a contour curve with contour points
hull = ConvexHull(landmarks)
# draw landmarks
lm = np.array(im)
for i in range(landmarks.shape[0]):
rr, cc = draw.circle(height-landmarks[i,1].astype('int32'), landmarks[i,0].astype('int32'), 5)
lm[rr, cc, :] = np.array((255, 0, 0))
# create mask
mask = np.zeros((height, width))
rr, cc = draw.polygon(height-landmarks[hull.vertices,1], landmarks[hull.vertices,0], mask.shape)
mask[rr,cc] = 1
path = path[:-1] if path[:-1] == '/' else path
image_name = path[path.rindex('/')+1:-4] + '_contour.png'
show_result(lm, mask, np.tile((mask!=0)[:,:,np.newaxis], (1,1,3)) * im, save=True, filename='images/'+image_name)
# add CRF
prob = np.concatenate(((1-mask)[np.newaxis,:,:]*0.9 +
mask[np.newaxis, :, :]*0.1,
mask[np.newaxis, :, :]*0.9 +
(1-mask)[np.newaxis, :, :]*0.1), axis=0)
map = CRF(prob, np.array(im))
def main(args):
image_paths = read_list(args.image_list)
for path in image_paths:
im = open_image(path)
# resize for memory
width, height = im.size
if height > 800:
im = im.resize((int(800 * width / height), 800))
width, height = im.size
# use 2D-FAN detect landmarks
fa = FaceAlignment(LandmarksType._2D,
enable_cuda=True,
flip_input=False,
use_cnn_face_detector=True)
try:
landmarks = fa.get_landmarks(np.array(im))[-1]
landmarks[:, 1] = height - landmarks[:, 1]
except:
continue
# generate a contour curve with contour points
hull = ConvexHull(landmarks)
# draw landmarks
lm = np.array(im)
for i in range(landmarks.shape[0]):
rr, cc = draw.circle(height - landmarks[i, 1].astype('int32'),
landmarks[i, 0].astype('int32'), 5)
lm[rr, cc, :] = np.array((255, 0, 0))
# create mask
mask = np.zeros((height, width))
rr, cc = draw.polygon(height - landmarks[hull.vertices, 1],
landmarks[hull.vertices, 0], mask.shape)
mask[rr, cc] = 1
save = True if args.save == 'True' else False
path = path[:-1] if path[:-1] == '/' else path
image_name = path[path.rindex('/') + 1:-4] + '_contour_nocrf.png'
show_result(lm,
mask,
np.tile((mask != 0)[:, :, np.newaxis], (1, 1, 3)) * im,
save=save,
filename='images/' + image_name)
# add CRF
#prob = np.concatenate(((1-mask)[np.newaxis,:,:]*0.9 + mask[np.newaxis, :, :]*0.1, mask[np.newaxis, :, :]*0.9 + (1-mask)[np.newaxis, :, :]*0.1), axis=0)
prob = ndimage.gaussian_filter(mask * 1.0, sigma=5)
prob = np.concatenate(
((1 - prob)[np.newaxis, :, :], prob[np.newaxis, :, :]), axis=0)
map = CRF(prob, np.array(im))
image_name = path[path.rindex('/') + 1:-4] + '_contour_crf.png'
show_result(im,
map,
np.tile((map != 0)[:, :, np.newaxis], (1, 1, 3)) * im,
save=save,
filename='images/' + image_name)
G_losses.append(G_train_loss.data)
num_iter += 1
epoch_end_time = time.time()
per_epoch_ptime = epoch_end_time - epoch_start_time
print('[%d/%d] - ptime: %.2f, loss_d: %.3f, loss_g: %.3f' %
((epoch + 1), opt.train_epoch, per_epoch_ptime,
torch.mean(torch.FloatTensor(D_losses)),
torch.mean(torch.FloatTensor(G_losses))))
fixed_p = root + 'Fixed_results/' + model + str(epoch + 1) + '.png'
util.show_result(G,
Variable(fixed_x_, volatile=True),
fixed_y_, (epoch + 1),
save=True,
path=fixed_p)
train_hist['per_epoch_ptimes'].append(per_epoch_ptime)
end_time = time.time()
total_ptime = end_time - start_time
train_hist['total_ptime'].append(total_ptime)
print("Avg one epoch ptime: %.2f, total %d epochs ptime: %.2f" %
(torch.mean(torch.FloatTensor(
train_hist['per_epoch_ptimes'])), opt.train_epoch, total_ptime))
print("Training finish!... save training results")
torch.save(G.state_dict(), root + model + 'generator_param.pkl')
torch.save(D.state_dict(), root + model + 'discriminator_param.pkl')
with open(root + model + 'train_hist.pkl', 'wb') as f:
G_losses.append(loss_g_)
train_hist['G_losses'].append(loss_g_)
num_iter += 1
epoch_end_time = time.time()
per_epoch_ptime = epoch_end_time - epoch_start_time
print('[%d/%d] - ptime: %.2f, loss_d: %.3f, loss_g: %.3f' %
((epoch + 1), opt.train_epoch, per_epoch_ptime, np.mean(
(D_losses)), np.mean(G_losses)))
fixed_p = root + 'Fixed_results/' + model + str(epoch + 1) + '.png'
outputs = sess.run(G, {x: fixed_x_})
util.show_result(fixed_x_,
outputs,
fixed_y_, (epoch + 1),
save=True,
path=fixed_p)
train_hist['per_epoch_ptimes'].append(per_epoch_ptime)
end_time = time.time()
total_ptime = end_time - start_time
train_hist['total_ptime'].append(total_ptime)
print("Avg. one epoch ptime: %.2f, total %d epochs ptime: %.2f" %
(np.mean(train_hist['per_epoch_ptimes']), opt.train_epoch, total_ptime))
print("Training finish!... save training results")
with open(root + model + 'train_hist.pkl', 'wb') as f:
pickle.dump(train_hist, f)
saver.save(sess, root + model + 'params.ckpt')
G_losses.append(G_train_loss.data)
num_iter += 1
epoch_end_time = time.time()
per_epoch_ptime = epoch_end_time - epoch_start_time
print('[%d/%d] - ptime: %.2f, loss_d: %.3f, loss_g: %.3f' %
((epoch + 1), opt.train_epoch, per_epoch_ptime,
torch.mean(torch.FloatTensor(D_losses)),
torch.mean(torch.FloatTensor(G_losses))))
fixed_p = root + 'Fixed_results/' + model + str(epoch + 1) + '.png'
util.show_result(G,
Variable(fixed_x_.cuda()),
fixed_y_, (epoch + 1),
save=True,
path=fixed_p)
train_hist['per_epoch_ptimes'].append(per_epoch_ptime)
end_time = time.time()
total_ptime = end_time - start_time
train_hist['total_ptime'].append(total_ptime)
print("Avg one epoch ptime: %.2f, total %d epochs ptime: %.2f" %
(torch.mean(torch.FloatTensor(
train_hist['per_epoch_ptimes'])), opt.train_epoch, total_ptime))
print("Training finish!... save training results")
torch.save(G.state_dict(), root + model + 'generator_param.pkl')
torch.save(D.state_dict(), root + model + 'discriminator_param.pkl')
with open(root + model + 'train_hist.pkl', 'wb') as f: