Can it work with a single image?

[aligoglos]

I wrote simple code to run model on a single image but result is gray still !!
minimal demo :

 import torch
import torch.nn as nn
from torchvision import transforms
from torchvision.utils import save_image
import cv2
from PIL import Image
import numpy as np
from tqdm import tqdm
import os
import argparse
import subprocess
import utils
import glob


def main():
	device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
	refimgs = None
	disable_colorization = False
	# Load remaster network
	modelR = __import__( 'model.remasternet', fromlist=['NetworkR'] ).NetworkR()
	state_dict = torch.load( 'remasternet.pth' )
	modelR.load_state_dict( state_dict['modelR'] )
	modelR = modelR.to(device)
	modelR.eval()
	if not disable_colorization:
		modelC = __import__( 'model.remasternet', fromlist=['NetworkC'] ).NetworkC()
		modelC.load_state_dict( state_dict['modelC'] )
		modelC = modelC.to(device)
		modelC.eval()
	paths = sorted(glob.glob('./inputs' + '/*'))
	for path in paths:
		image = cv2.imread(path)
		if ~(image is None):
			name = path.split('\\')[-1]
			print(name)
			refimgs = cv2.imread(F"./references/{name}")
			with torch.no_grad():
				gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
				frame_l = torch.from_numpy(gray).view( gray.shape[0], gray.shape[1], 1 )
				frame_l = frame_l.permute(2, 0, 1).float() # HWC to CHW
				frame_l /= 255.
				frame_l = frame_l.view(1, frame_l.size(0), 1, frame_l.size(1), frame_l.size(2))
				input = frame_l.to( device )
				output_l = modelR( input )
				if refimgs is None:
					output_ab = modelC( output_l )
				else:
					refimgs = torch.from_numpy(refimgs)
					refimgs = refimgs.permute(2, 0, 1).float().unsqueeze(axis = 0).unsqueeze(axis = 0)
					refimgs /= 255.
					refimgs = refimgs.to( device )
					output_ab = modelC( output_l, refimgs )

				output_l = output_l.detach().cpu()
				output_ab = output_ab.detach().cpu()
				out_l = output_l[0,:,0,:,:]
				out_c = output_ab[0,:,0,:,:]
				output = torch.cat((out_l, out_c), dim=0).numpy().transpose((1, 2, 0))
				output = Image.fromarray( np.uint8( utils.convertLAB2RGB( output )*255 ) )
				output.save( F"./results/{name}" )

if __name__ == "__main__":
	main()

input image :

out put :

** Note : reference image is equal to input

[zhaoyuzhi]

I also encounter the same issue (output grayscale when single frame as input). Have you addressed this issue?

[hermosayhl]

I encounter this issue, too. Is there anyone make it ?

[hermosayhl]

I also encounter the same issue (output grayscale when single frame as input). Have you addressed this issue?

Dr zhao, have you overcome this problem?

[Dawars]

You have to emulate multiple frames by duplicating the image to make the temporal convolutions work:

input = torch.tile(input, (1, 1, 5, 1, 1))

The network still isn't able to use colors from the reference images if they are significantly different from the gray image.