github/DeepFaceLab
1
0
Fork 0
mirror of https://github.com/iperov/DeepFaceLab.git synced 2025-07-06 21:12:07 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

refactoring. Added RecycleGAN for testing.

Browse source
This commit is contained in:
iperov 2018-12-28 19:38:52 +04:00
parent 8686309417
commit f8824f9601
24 changed files with 1661 additions and 1505 deletions
Download patch file Download diff file Expand all files Collapse all files

103
utils/image_utils.py
View file

@ -5,6 +5,7 @@ import cv2
import localization
from scipy.spatial import Delaunay
from PIL import Image, ImageDraw, ImageFont
from nnlib import nnlib
def channel_hist_match(source, template, hist_match_threshold=255, mask=None):
# Code borrowed from:
@ -276,103 +277,21 @@ def reduce_colors (img_bgr, n_colors):
class TFLabConverter():
def __init__(self,):
import gpufmkmgr
def __init__(self):
exec (nnlib.import_tf(), locals(), globals())
self.tf_sess = tf_sess
self.tf_module = gpufmkmgr.import_tf()
self.tf_session = gpufmkmgr.get_tf_session()
self.bgr_input_tensor = tf.placeholder("float", [None, None, 3])
self.lab_input_tensor = tf.placeholder("float", [None, None, 3])
self.bgr_input_tensor = self.tf_module.placeholder("float", [None, None, 3])
self.lab_input_tensor = self.tf_module.placeholder("float", [None, None, 3])
self.lab_output_tensor = self.rgb_to_lab(self.tf_module, self.bgr_input_tensor)
self.bgr_output_tensor = self.lab_to_rgb(self.tf_module, self.lab_input_tensor)
self.lab_output_tensor = tf_rgb_to_lab()(self.bgr_input_tensor)
self.bgr_output_tensor = tf_lab_to_rgb()(self.lab_input_tensor)
def bgr2lab(self, bgr):
return self.tf_session.run(self.lab_output_tensor, feed_dict={self.bgr_input_tensor: bgr})
return self.tf_sess.run(self.lab_output_tensor, feed_dict={self.bgr_input_tensor: bgr})
def lab2bgr(self, lab):
return self.tf_session.run(self.bgr_output_tensor, feed_dict={self.lab_input_tensor: lab})
return self.tf_sess.run(self.bgr_output_tensor, feed_dict={self.lab_input_tensor: lab})
def rgb_to_lab(self, tf, rgb_input):
with tf.name_scope("rgb_to_lab"):
srgb_pixels = tf.reshape(rgb_input, [-1, 3])
with tf.name_scope("srgb_to_xyz"):
linear_mask = tf.cast(srgb_pixels <= 0.04045, dtype=tf.float32)
exponential_mask = tf.cast(srgb_pixels > 0.04045, dtype=tf.float32)
rgb_pixels = (srgb_pixels / 12.92 * linear_mask) + (((srgb_pixels + 0.055) / 1.055) ** 2.4) * exponential_mask
rgb_to_xyz = tf.constant([
# X Y Z
[0.412453, 0.212671, 0.019334], # R
[0.357580, 0.715160, 0.119193], # G
[0.180423, 0.072169, 0.950227], # B
])
xyz_pixels = tf.matmul(rgb_pixels, rgb_to_xyz)
# https://en.wikipedia.org/wiki/Lab_color_space#CIELAB-CIEXYZ_conversions
with tf.name_scope("xyz_to_cielab"):
# convert to fx = f(X/Xn), fy = f(Y/Yn), fz = f(Z/Zn)
# normalize for D65 white point
xyz_normalized_pixels = tf.multiply(xyz_pixels, [1/0.950456, 1.0, 1/1.088754])
epsilon = 6/29
linear_mask = tf.cast(xyz_normalized_pixels <= (epsilon**3), dtype=tf.float32)
exponential_mask = tf.cast(xyz_normalized_pixels > (epsilon**3), dtype=tf.float32)
fxfyfz_pixels = (xyz_normalized_pixels / (3 * epsilon**2) + 4/29) * linear_mask + (xyz_normalized_pixels ** (1/3)) * exponential_mask
# convert to lab
fxfyfz_to_lab = tf.constant([
# l a b
[ 0.0, 500.0, 0.0], # fx
[116.0, -500.0, 200.0], # fy
[ 0.0, 0.0, -200.0], # fz
])
lab_pixels = tf.matmul(fxfyfz_pixels, fxfyfz_to_lab) + tf.constant([-16.0, 0.0, 0.0])
return tf.reshape(lab_pixels, tf.shape(rgb_input))
def lab_to_rgb(self, tf, lab):
with tf.name_scope("lab_to_rgb"):
lab_pixels = tf.reshape(lab, [-1, 3])
# https://en.wikipedia.org/wiki/Lab_color_space#CIELAB-CIEXYZ_conversions
with tf.name_scope("cielab_to_xyz"):
# convert to fxfyfz
lab_to_fxfyfz = tf.constant([
# fx fy fz
[1/116.0, 1/116.0, 1/116.0], # l
[1/500.0, 0.0, 0.0], # a
[ 0.0, 0.0, -1/200.0], # b
])
fxfyfz_pixels = tf.matmul(lab_pixels + tf.constant([16.0, 0.0, 0.0]), lab_to_fxfyfz)
# convert to xyz
epsilon = 6/29
linear_mask = tf.cast(fxfyfz_pixels <= epsilon, dtype=tf.float32)
exponential_mask = tf.cast(fxfyfz_pixels > epsilon, dtype=tf.float32)
xyz_pixels = (3 * epsilon**2 * (fxfyfz_pixels - 4/29)) * linear_mask + (fxfyfz_pixels ** 3) * exponential_mask
# denormalize for D65 white point
xyz_pixels = tf.multiply(xyz_pixels, [0.950456, 1.0, 1.088754])
with tf.name_scope("xyz_to_srgb"):
xyz_to_rgb = tf.constant([
# r g b
[ 3.2404542, -0.9692660, 0.0556434], # x
[-1.5371385, 1.8760108, -0.2040259], # y
[-0.4985314, 0.0415560, 1.0572252], # z
])
rgb_pixels = tf.matmul(xyz_pixels, xyz_to_rgb)
# avoid a slightly negative number messing up the conversion
rgb_pixels = tf.clip_by_value(rgb_pixels, 0.0, 1.0)
linear_mask = tf.cast(rgb_pixels <= 0.0031308, dtype=tf.float32)
exponential_mask = tf.cast(rgb_pixels > 0.0031308, dtype=tf.float32)
srgb_pixels = (rgb_pixels * 12.92 * linear_mask) + ((rgb_pixels ** (1/2.4) * 1.055) - 0.055) * exponential_mask
return tf.reshape(srgb_pixels, tf.shape(lab))