Borderline ovarian neoplasms

Influence of Color Spaces for Deep Learning Image Colorization

“…DeOldify [Antic, 2019] is another end-to-end image and video colorization method mapping the missing chrominance values to the grayscale input image. A ResNet (ResNet101 or ResNet34) is used as the backbone of the generator of a U-Net architecture trained as follows : the generator is first trained with the perceptual loss [Johnson et al, 2016], followed by training the critic as a binary classsifier distinguishing between real images and those generated by the generator and finally the generator and critic are trained together in an adversarial manner on 1 − 3% of the ImageNet [Deng et al, 2009] data.…”

“…Histogram User Diverse Object Survey GANs prediction guided aware [Cheng et al, 2015] [Iizuka et al, 2016] [Vitoria et al, 2020] [Nazeri et al, 2018] [Cao et al, 2017] [Yoo et al, 2019] [Antic, 2019] [Larsson et al, 2016] [Zhang et al, 2016] [Mouzon et al, 2019] [He et al, 2018] [Deshpande et al, 2017] [Guadarrama et al, 2017] [Royer et al, 2017] [Kumar et al, 2021] [Su et al, 2020] [Pucci et al, 2021 [ Kong et al, 2021] [ Gu et al, 2019] (winner)…”

“…Very few papers in the literature tackle old black and white images' colorization. In deep learning-based approaches, [Vitoria et al, 2020, Antic, 2019 present some results on Legacy Black and White Photographs while [Luo et al, 2020] restores and colorizes old black and white portraits. [Wan et al, 2020b] focuses on the restoration of old photos by training two variational autoencoders (VAE) to project clean and old photos to two latent spaces and to learn the translation between these latent spaces on synthetic paired data.…”