def main():
# paths
data_dir = "../data/"
# load images
imgs = []
paths = [data_dir + "./lenna.jpg", data_dir + "./cat.jpg"]
for i in range(len(paths)):
img = img2array(paths[i], desired_size=[512, 512], expand=True)
imgs.append(torch.from_numpy(img))
imgs = torch.cat(imgs).permute(0, 3, 1, 2)
# loc = torch.Tensor(2, 2).uniform_(-1, 1)
loc = torch.from_numpy(np.array([[0.0, 0.0], [0.0, 0.0]]))
num_patches = 5
scale = 2
patch_size = 10
ret = Retina(g=patch_size, k=num_patches, s=scale)
glimpse = ret.foveate(imgs, loc).data.numpy()
glimpse = np.reshape(glimpse, [2, num_patches, 3, patch_size, patch_size])
glimpse = np.transpose(glimpse, [0, 1, 3, 4, 2])
merged = []
for i in range(len(glimpse)):
g = glimpse[i]
g = list(g)
g = [array2img(l) for l in g]
res = reduce(merge_images, list(g))
merged.append(res)
merged = [np.asarray(l, dtype="float32") / 255.0 for l in merged]
fig, axs = plt.subplots(nrows=2, ncols=1)
for i, ax in enumerate(axs.flat):
axs[i].imshow(merged[i])
axs[i].get_xaxis().set_visible(False)
axs[i].get_yaxis().set_visible(False)
plt.show()
def main():
# load images
imgs = []
paths = [data_dir + './lenna.jpg', data_dir + './cat.jpg']
for i in range(len(paths)):
img = img2array(paths[i], desired_size=[512, 512], expand=True)
imgs.append(torch.from_numpy(img))
imgs = Variable(torch.cat(imgs))
imgs = imgs.permute(0, 3, 1, 2)
# loc = torch.Tensor(2, 2).uniform_(-1, 1)
loc = torch.from_numpy(np.array([[0., 0.], [0., 0.]]))
loc = Variable(loc)
ret = retina(g=64, k=3, s=2)
glimpse = ret.foveate(imgs, loc).data.numpy()
glimpse = np.reshape(glimpse, [2, 3, 3, 64, 64])
glimpse = np.transpose(glimpse, [0, 1, 3, 4, 2])
merged = []
for i in range(len(glimpse)):
g = glimpse[i]
g = list(g)
g = [array2img(l) for l in g]
res = reduce(merge_images, list(g))
merged.append(res)
merged = [np.asarray(l, dtype='float32') / 255.0 for l in merged]
fig, axs = plt.subplots(nrows=2, ncols=1)
for i, ax in enumerate(axs.flat):
axs[i].imshow(merged[i])
axs[i].get_xaxis().set_visible(False)
axs[i].get_yaxis().set_visible(False)
plt.show()
train_data = utils.readImg2array(train_path,
size,
threshold,
pflag=pflag,
nflag=nflag)
# create a list of tese data, e.g. yosukekatada 's sunglasses picture.
test_path = os.path.join(cwd_path, 'data/sunglasses.jpg')
test_data = utils.readImg2array(test_path,
size,
threshold,
pflag=pflag,
nflag=nflag)
# build model
model = dhnn.DHNN(pflag=pflag, nflag=nflag)
print('[START] training.')
model.train([train_data.flatten()])
print('[END] training.')
print('[START] predicting.')
recovery = model.predict([test_data.flatten()], epochs=50000)
print('[END] predicting.')
print('[START] predicting.')
recovery = recovery[0].reshape(size)
outfile = os.path.join(cwd_path, 'data', 'recovery.jpg')
utils.array2img(recovery, outFile=outfile, pflag=pflag, nflag=nflag)
print('[END] saving.')
input_img = np.concatenate([img1, img2, img3, img4], axis=0)
B, H, W, C = input_img.shape
print("Input Img Shape: {}".format(input_img.shape))
# identity transform
theta = np.array([[1., 0, 0], [0, 1., 0]])
x = tf.placeholder(tf.float32, [None, H, W, C])
with tf.variable_scope('spatial_transformer'):
theta = theta.astype('float32')
theta = theta.flatten()
# define loc net weight and bias
loc_in = H * W * C
loc_out = 6
W_loc = tf.Variable(tf.zeros([loc_in, loc_out]), name='W_loc')
b_loc = tf.Variable(initial_value=theta, name='b_loc')
# tie everything together
fc_loc = tf.matmul(tf.zeros([B, loc_in]), W_loc) + b_loc
h_trans = transformer(x, fc_loc)
# run session
sess = tf.Session()
sess.run(tf.global_variables_initializer())
y = sess.run(h_trans, feed_dict={x: input_img})
print("y: {}".format(y.shape))
array2img(y[0]).show()
newData = subImgs[:, idx]
if data == None:
data = newData
else:
data = numpy.hstack((data, newData))
if i % actualisation == 0:
U, D, mu, n = skl(data=data,
U0=U,
D0=D,
mu0=mu,
n0=n,
ff=ff,
K=K)
data = None
cv.ShowImage("Warped Stream", array2img(subImgs[:, idx], outSize))
cv.ShowImage("Mean Stream", array2img(mu, outSize))
cv.ShowImage("1st Eigenface Stream", array2img(U[:, 0], outSize))
i = i + 1
# Draw the extraction polygon
cv.PolyLine(inImgGray, (polygon.corners(), ), True,
cv.RGB(255, 255, 255))
cv.ShowImage("Webcam Stream", inImgGray)
keyPressed = cv.WaitKey(10)
if keyPressed == 10:
extract = not (extract)
elif keyPressed != -1:
break
b_loc = tf.Variable(initial_value=theta, name='b_loc')
# tie everything together
fc_loc = tf.matmul(tf.zeros([B, loc_in]), W_loc) + b_loc
h_trans = transformer(x, fc_loc)
# inverse
inv_theta = inv_theta.astype('float32')
inv_theta = inv_theta.flatten()
# define loc net weight and bias
loc_in = H * W * C
loc_out = 6
W_loc = tf.Variable(tf.zeros([loc_in, loc_out]), name='W_loc')
b_loc = tf.Variable(initial_value=inv_theta, name='b_loc')
# tie everything together
fc_loc = tf.matmul(tf.zeros([B, loc_in]), W_loc) + b_loc
inv_h_trans = transformer(x, fc_loc)
# run session
sess = tf.Session()
sess.run(tf.global_variables_initializer())
y = sess.run(h_trans, feed_dict={x: input_img})
print("y: {}".format(y.shape))
array2img(y[0]).show()
out_img = sess.run(inv_h_trans, feed_dict={x: y})
print("inv: {}".format(out_img.shape))
array2img(out_img[0]).show()
subImgs = numpy.mat(cv2array(subImgTmp)).ravel().T
else:
subImgs = numpy.hstack((subImgs, numpy.mat(cv2array(subImgTmp)).ravel().T))
idx = maxLikelihood(subImgs, U, D, mu)
polygon = polygonTmp[idx]
# SKL!!!
newData = subImgs[:, idx]
if data == None:
data = newData
else:
data = numpy.hstack((data, newData))
if i % actualisation == 0:
U, D, mu, n = skl(data=data, U0=U, D0=D, mu0=mu, n0=n, ff=ff, K=K)
data = None
cv.ShowImage("Warped Stream", array2img(subImgs[:, idx], outSize))
cv.ShowImage("Mean Stream", array2img(mu, outSize))
cv.ShowImage("1st Eigenface Stream", array2img(U[:, 0], outSize))
i = i + 1
# Draw the extraction polygon
cv.PolyLine(inImgGray, (polygon.corners(),), True, cv.RGB(255, 255, 255))
cv.ShowImage("Webcam Stream", inImgGray)
keyPressed = cv.WaitKey(10)
if keyPressed == 10:
extract = not(extract)
elif keyPressed != -1:
break