File size: 7,194 Bytes
0caed3c 0004858 0caed3c df23063 0caed3c df23063 0caed3c df23063 0caed3c df23063 0caed3c df23063 0caed3c df23063 |
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 |
import os
import argparse
import numpy as np
from skimage import color, io
import torch
import torch.nn.functional as F
from PIL import Image
from models import ColorEncoder, ColorUNet
from extractor.manga_panel_extractor import PanelExtractor
os.environ["CUDA_VISIBLE_DEVICES"] = '0'
def mkdirs(path):
if not os.path.exists(path):
os.makedirs(path)
def Lab2RGB_out(img_lab):
img_lab = img_lab.detach().cpu()
img_l = img_lab[:,:1,:,:]
img_ab = img_lab[:,1:,:,:]
img_l = img_l + 50
pred_lab = torch.cat((img_l, img_ab), 1)[0,...].numpy()
out = (np.clip(color.lab2rgb(pred_lab.transpose(1, 2, 0)), 0, 1)* 255).astype("uint8")
return out
def RGB2Lab(inputs):
return color.rgb2lab(inputs)
def Normalize(inputs):
l = inputs[:, :, 0:1]
ab = inputs[:, :, 1:3]
l = l - 50
lab = np.concatenate((l, ab), 2)
return lab.astype('float32')
def numpy2tensor(inputs):
out = torch.from_numpy(inputs.transpose(2,0,1))
return out
def tensor2numpy(inputs):
out = inputs[0,...].detach().cpu().numpy().transpose(1,2,0)
return out
def preprocessing(inputs):
img_lab = Normalize(RGB2Lab(inputs))
img = np.array(inputs, 'float32')
img = numpy2tensor(img)
img_lab = numpy2tensor(img_lab)
return img.unsqueeze(0), img_lab.unsqueeze(0)
if __name__ == "__main__":
device = "cuda"
parser = argparse.ArgumentParser()
parser.add_argument("--path", type=str, default=None, help="path of input image")
parser.add_argument("--size", type=int, default=None)
parser.add_argument("--ckpt", type=str, default=None, help="path of model weight")
parser.add_argument("-ne", "--no_extractor", action='store_true', help="Do not segment the manga panels.")
args = parser.parse_args()
if args.path:
test_dir_path = args.path
if args.size:
imgsize = args.size
if args.ckpt:
ckpt_path = args.ckpt
if args.no_extractor:
no_extractor = args.no_extractor
ckpt = torch.load(ckpt_path, map_location=lambda storage, loc: storage)
colorEncoder = ColorEncoder().to(device)
colorEncoder.load_state_dict(ckpt["colorEncoder"])
colorEncoder.eval()
colorUNet = ColorUNet().to(device)
colorUNet.load_state_dict(ckpt["colorUNet"])
colorUNet.eval()
imgs = []
imgs_lab = []
while 1:
print(f'make sure both manga image and reference images are under this path {test_dir_path}')
img_path = input("please input the name of image needed to be colorized (with file extension): ")
img_path = os.path.join(test_dir_path, img_path)
img_name = os.path.basename(img_path)
img_name = os.path.splitext(img_name)[0]
if no_extractor:
ref_img_path = os.path.join(test_dir_path, input(f"Enter the reference image path: "))
img1 = Image.open(img_path).convert("RGB")
width, height = img1.size
img2 = Image.open(ref_img_path).convert("RGB")
img1, img1_lab = preprocessing(img1)
img2, img2_lab = preprocessing(img2)
img1 = img1.to(device)
img1_lab = img1_lab.to(device)
img2 = img2.to(device)
img2_lab = img2_lab.to(device)
with torch.no_grad():
img2_resize = F.interpolate(img2 / 255., size=(imgsize, imgsize), mode='bilinear',
recompute_scale_factor=False, align_corners=False)
img1_L_resize = F.interpolate(img1_lab[:, :1, :, :] / 50., size=(imgsize, imgsize), mode='bilinear',
recompute_scale_factor=False, align_corners=False)
color_vector = colorEncoder(img2_resize)
fake_ab = colorUNet((img1_L_resize, color_vector))
fake_ab = F.interpolate(fake_ab * 110, size=(height, width), mode='bilinear',
recompute_scale_factor=False, align_corners=False)
fake_img = torch.cat((img1_lab[:, :1, :, :], fake_ab), 1)
fake_img = Lab2RGB_out(fake_img)
out_folder = os.path.dirname(img_path)
out_name = os.path.basename(img_path)
out_name = os.path.splitext(out_name)[0]
out_img_path = os.path.join(out_folder, 'color', f'{out_name}_color.png')
# show image
Image.fromarray(fake_img).show()
# save image
folder_path = os.path.join(out_folder, 'color')
if not os.path.exists(folder_path):
os.makedirs(folder_path)
io.imsave(out_img_path, fake_img)
continue
panel_extractor = PanelExtractor(min_pct_panel=5, max_pct_panel=90)
panels, masks, panel_masks = panel_extractor.extract(img_path)
panel_num = len(panels)
ref_img_paths = []
print("Please enter the name of the reference image in order according to the number prompts on the picture")
for i in range(panel_num):
ref_img_path = os.path.join(test_dir_path, input(f"{i+1}/{panel_num} reference image:"))
ref_img_paths.append(ref_img_path)
fake_imgs = []
for i in range(panel_num):
img1 = Image.fromarray(panels[i]).convert("RGB")
width, height = img1.size
img2 = Image.open(ref_img_paths[i]).convert("RGB")
img1, img1_lab = preprocessing(img1)
img2, img2_lab = preprocessing(img2)
img1 = img1.to(device)
img1_lab = img1_lab.to(device)
img2 = img2.to(device)
img2_lab = img2_lab.to(device)
with torch.no_grad():
img2_resize = F.interpolate(img2 / 255., size=(imgsize, imgsize), mode='bilinear', recompute_scale_factor=False, align_corners=False)
img1_L_resize = F.interpolate(img1_lab[:,:1,:,:] / 50., size=(imgsize, imgsize), mode='bilinear', recompute_scale_factor=False, align_corners=False)
color_vector = colorEncoder(img2_resize)
fake_ab = colorUNet((img1_L_resize, color_vector))
fake_ab = F.interpolate(fake_ab*110, size=(height, width), mode='bilinear', recompute_scale_factor=False, align_corners=False)
fake_img = torch.cat((img1_lab[:,:1,:,:], fake_ab), 1)
fake_img = Lab2RGB_out(fake_img)
fake_imgs.append(fake_img)
if panel_num == 1:
out_folder = os.path.dirname(img_path)
out_name = os.path.basename(img_path)
out_name = os.path.splitext(out_name)[0]
out_img_path = os.path.join(out_folder,'color',f'{out_name}_color.png')
Image.fromarray(fake_imgs[0]).show()
folder_path = os.path.join(out_folder, 'color')
if not os.path.exists(folder_path):
os.makedirs(folder_path)
io.imsave(out_img_path, fake_imgs[0])
else:
panel_extractor.concatPanels(img_path, fake_imgs, masks, panel_masks)
print(f'Colored images have been saved to: {os.path.join(test_dir_path, "color")}')
|