“An Analysis: Different Methods about Line Art Colorization”, Jinhui Gao, Ruihao Zeng, Yuan Liang, Xinyu Diao2022-11-10 (Danbooru AI, GAN)⁠:

We have conducted a series of studies and analyses to address the problem of line art colorization. We chose Generative Adversarial Networks (GANs), a leading neural network architecture for solving this problem, as our focus.

For a large number of studies based on this architecture, we improved, applied, and analytically compared 4 methods, pix2pix, pix2pixHD, white-box, and scaled Fourier transform (SCFT), which can represent the mainstream problem-solving direction in the field of line colorization to the greatest extent possible.

Finally, two reference quantities were introduced to quantify the results of the analysis…From the coloring results and the two indicators, we can see that the white-box and pix2pixHD methods are relatively good coloring results, and pix2pix is less effective.

…4.1 Dataset description: We use a dataset that consists of richly tagged and labeled artwork depicting characters from Japanese anime, and they are collected from two imageboards Danbooru and Moeimouto. All images in the dataset have been tagged as SFW (non-explicit). The dataset file has a subset of 300,000 images that are in normalized size format of 512 × 512 px. The total amount of data is about 45.34 GB. In this study, we only took some data for experiments.