Lev Manovich and Jeremy Douglass, 2010.
883 Manga series from the scanlation site OneManga.com.
Total number of pages: 1,074,790.
In the Fall 2009, we have downloaded 883 Manga series containing 1,074,790 unique pages from this site. We have then used our custom software system installed on a supercomputer at National Department of Energy Research Center (NERSC) to analyze visual features of these pages.
The longest running Manga series has been published continuously since 1976. The most popular series on OneManga.com are Naruto (1999-; 8835 pages) and One Piece (1997-; 10562 pages). Along with such long Manga series, our data set also contains shorter series that appeared in 2000s and only run for 1-3 years.
X axis: standard deviation of pixels’ grayscale values in a page.
Y axis: entropy measured over all pixels’ grayscale values in a page.
The visualization shows 1,074,790 unique pages from 883 distinct manga series from Japan, Korea and China. The series include both very popular long-running titles such as Naruto and One Piece and also many short-lived titles. The visualization maps the pages the pages according to some of their visual characteristics that were measured automatically on supercomputers at the U.S. National Department of Energy Research Center using custom software developed by Software Studies Initiative.
(X-axis: standard deviation. Y-axis: entropy.)
The pages in the bottom part of the visualization are the most graphic and have the least amount of detail. The pages in the upper right have lots of detail and texture. The pages with the highest contrast are on the right, while pages with the least contrast are on the left. In between these four extremes, we find every possible stylistic variation.
This suggests that our basic concept of “style” maybe not appropriate then we consider large cultural data sets. The concept assumes that we can partition a set of cultural artifacts works into a small number of discrete categories. In the case of our one million pages set, we find practically infinite graphical variations. If we try to divide this space into discrete stylistic categories, any such attempt will be arbitrary.
Visualization also shows which graphical choices are more commonly used by manga artists (the central part of the “cloud” of pages) and which appear much more rarely (bottom and left parts).
Note: some of the pages - such as all covers - are in color. However in order to be able to fit all image into a single large image (the original is 44,000x44,000 pixels - scaled to 10,000x10,000 for posting to Flickr), we rendered everything in grey scale.
Because pages are rendered on top of each other, you don't actually see 1 million of distinct pages - the visualization shows a distribution of all pages with typical examples appearing on the top.