def bce_loss(preds, targets):
"""
Calculates the sum of binary cross-entropy losses between predictions and ground truths.
@param preds: A 1xN tensor. The predicted classifications of each frame.
@param targets: A 1xN tensor The target labels for each frame. (Either 1 or -1). Not "truths"
because the generator passes in lies to determine how well it confuses the
discriminator.
@return: The sum of binary cross-entropy losses.
"""
return tf.squeeze(-1 * (tf.matmul(targets, log10(preds), transpose_a=True) +
tf.matmul(1 - targets, log10(1 - preds), transpose_a=True)))
def psnr_error(gen_frames, gt_frames):
"""
Computes the Peak Signal to Noise Ratio error between the generated images and the ground
truth images.
@param gen_frames: A tensor of shape [batch_size, height, width, 3]. The frames generated by the
generator model.
@param gt_frames: A tensor of shape [batch_size, height, width, 3]. The ground-truth frames for
each frame in gen_frames.
@return: A scalar tensor. The mean Peak Signal to Noise Ratio error over each frame in the
batch.
"""
shape = tf.shape(gen_frames)
num_pixels = tf.to_float(shape[1] * shape[2] * shape[3])
square_diff = tf.square(gt_frames - gen_frames)
batch_errors = 10 * log10(1 / ((1 / num_pixels) * tf.reduce_sum(square_diff, [1, 2, 3])))
return tf.reduce_mean(batch_errors)
def sharp_diff_error(gen_frames, gt_frames):
"""
Computes the Sharpness Difference error between the generated images and the ground truth
images.
@param gen_frames: A tensor of shape [batch_size, height, width, 3]. The frames generated by the
generator model.
@param gt_frames: A tensor of shape [batch_size, height, width, 3]. The ground-truth frames for
each frame in gen_frames.
@return: A scalar tensor. The Sharpness Difference error over each frame in the batch.
"""
shape = tf.shape(gen_frames)
num_pixels = tf.to_float(shape[1] * shape[2] * shape[3])
with tf.name_scope('grad_diff'):
# gradient difference
# create filters [-1, 1] and [[1],[-1]] for diffing to the left and down respectively.
with tf.name_scope('setup'):
pos = tf.constant(np.identity(3), dtype=tf.float32)
neg = -1 * pos
filter_x = tf.expand_dims(tf.stack([neg, pos]), 0) # [-1, 1]
filter_y = tf.stack([tf.expand_dims(pos, 0), tf.expand_dims(neg, 0)]) # [[1],[-1]]
strides = [1, 1, 1, 1] # stride of (1, 1)
padding = 'SAME'
with tf.name_scope('gen_dx'):
gen_dx = tf.abs(tf.nn.conv2d(gen_frames, filter_x, strides, padding=padding))
with tf.name_scope('gen_dy'):
gen_dy = tf.abs(tf.nn.conv2d(gen_frames, filter_y, strides, padding=padding))
with tf.name_scope('gt_dx'):
gt_dx = tf.abs(tf.nn.conv2d(gt_frames, filter_x, strides, padding=padding))
with tf.name_scope('gt_dy'):
gt_dy = tf.abs(tf.nn.conv2d(gt_frames, filter_y, strides, padding=padding))
gen_grad_sum = gen_dx + gen_dy
gt_grad_sum = gt_dx + gt_dy
grad_diff = tf.abs(gt_grad_sum - gen_grad_sum)
batch_errors = 10 * log10(1 / ((1 / num_pixels) * tf.reduce_sum(grad_diff, [1, 2, 3])))
return tf.reduce_mean(batch_errors)
def bce_loss(preds, targets):
return tf.squeeze(
-1 * (tf.matmul(targets, log10(preds), transpose_a=True) +
tf.matmul(1 - targets, log10(1 - preds), transpose_a=True)))
