def load_model():
print('load model')
device.set(device=DeviceId.GPU0)
warnings.filterwarnings("ignore", category=UserWarning, message=".*?Your .*? set is empty.*?")
model = get_image_colorizer(artistic=True)
#model = get_image_colorizer(artistic=False)
return model
def main():
torch.backends.cudnn.benchmark=True
#choices: CPU, GPU0...GPU7
device.set(device=DeviceId.GPU0)
if not torch.cuda.is_available():
print('GPU not available.')
warnings.filterwarnings("ignore", category=UserWarning, message=".*?Your .*? set is empty.*?")
colorizer = get_video_colorizer()
source_url = 'https://www.reddit.com/r/nextfuckinglevel/comments/jyq24w/this_shot_from_the_movie_wings_1927_is_too_good/' #@param {type:"string"}
render_factor = 21 #@param {type: "slider", min: 5, max: 40}
watermarked = False #@param {type:"boolean"}
if source_url is not None and source_url !='':
video_path = colorizer.colorize_from_url(source_url, 'video.mp4', render_factor, watermarked=watermarked)
else:
print('Provide a video url and try again.')
from deoldify import device
from deoldify.device_id import DeviceId
device.set(device=DeviceId.GPU0)
from deoldify.visualize import *
plt.style.use('dark_background')
import warnings
warnings.filterwarnings("ignore",
category=UserWarning,
message=".*?Your .*? set is empty.*?")
colorizer = get_video_colorizer()
def colorize(video_path: str, render_factor: int = 21):
# [5, 45] 21
#NOTE: Make source_url None to just read from file at ./video/source/[file_name] directly without modification
source_url = 'https://twitter.com/silentmoviegifs/status/1116751583386034176'
file_name = 'DogShy1926'
file_name_ext = file_name + '.mp4'
result_path = None
if video_path is not None:
result_path = colorizer.colorize_from_file_name(
video_path, render_factor=render_factor)
else:
print("No video_path provided, defaulting to a example picture.")
result_path = colorizer.colorize_from_url(source_url,
file_name_ext,
render_factor=render_factor)
help="input file path")
parser.add_argument("--out_file",
required=True,
type=str,
help="output file path")
parser.add_argument("--gpu",
required=True,
type=DeviceId.argparse,
choices=list(DeviceId))
parser.add_argument("--render_factor",
default=21,
type=int,
help="colorization render factor")
args = parser.parse_args()
in_file = args.in_file
out_file = args.out_file
gpu = args.gpu
render_factor = args.render_factor
device.set(device=gpu)
import fastai
from deoldify.visualize import *
from PIL import Image
print(f'Colorizing {in_file} to {out_file}')
colorizer = get_image_colorizer(artistic=False)
img = colorizer.get_transformed_image(in_file, render_factor=render_factor)
img.save(out_file)
