Two-stage pyramidal convolutional neural networks for image colorization

Abstract

The development of colorization algorithms through deep learning has become the current research trend. These algorithms colorize grayscale images automatically and quickly, but the colors produced are usually subdued and have low saturation. This research addresses this issue of existing algorithms by presenting a two-stage convolutional neural network (CNN) structure with the first and second stages being a chroma map generation network and a refinement network, respectively. To begin, we convert the color space of an image from RGB to HSV to predict its low-resolution chroma components and therefore reduce the computational complexity. Following that, the first-stage output is zoomed in and its detail is enhanced with a pyramidal CNN, resulting in a colorized image. Experiments show that, while using fewer parameters, our methodology produces results with more realistic color and higher saturation than existing methods.

Robust Manga Page Colorization via Coloring Latent Space

“…Therefore, we use StyleGAN to generate very rough colorings and to provide latent space specifically for these colorings. [39] employs rough colorings for colorization and shows their effectiveness. The model for rough coloring generation is also trained with adversarial loss, so the constraint on color diversity remains.…”