Software automating manga production
In the 21st Century, soft power seems to be more useful and effective than hard power in terms of shaping international influence. In fact, pop culture has become the most powerful tool to build a country’s soft power which in turn generates huge volumes of cultural and commodity exports.
The cool factor of Japanese culture might from the long-established manga and anime tradition – comics and animation targeting people of all ages, in genres as varied as office politics and historical dramas.
To help entrepreneurs tap into the huge potential of the most happening pop medium, researchers at The Chinese University of Hong Kong (CUHK) combined their expertise in visual perception and computation to automate the production of manga, including instant extraction of speech balloons, conversion of manga background from photographs, and colouring black and white manga intelligently.
CUHK is now looking for licensees from around the world who are interested in capitalising the commercial potential of computational manga technologies.
Exploiting the thriving pop culture
Manga is one of the most flourishing cultural and creative industries in the 21st Century. In 2013, the market grew eight per cent year-on-year to around US$70 million in the US and Canada.
Traditionally manga or comics are heavily based on the personal style of the artist. As time goes by, many comic artists nowadays use photos and other modern tools to help streamline production. Hence, a research team at CUHK devised three different technologies related to manga automation.
First off, famous manga and comics are commonly licensed in various countries and translated into different languages. The research team invented a programme that extracts speech balloons instantly from manga with precise boundaries that don’t need to be cleaned up manually (error less than one pixel on average). Coupled with translation software and multilingual support, the programme speeds up manga translation in major ways.
The increasing popularity of reading manga on portable devices such as tablet and smartphones has changed the landscape of the manga industry. Typically, authentic Japanese manga are in black and white, but there is a growing demand for coloured content on electronic screens.
The second technology the research team came up with is an automated context-aware manga colourisation programme based on pattern and intensity continuity, allowing the artist to colour artwork conveniently and quickly. Just scribble on the areas in desired colours and the programme will intelligently interpret where one colour stops and the other starts.
Finally, drawing background for manga can be tedious and labour intensive. The research team thus developed a programme that generates manga line drawings with exquisite half-tone pattern textures from colour photos. It preserves the visual richness of the original by utilising both screen density and patterns whereas the mapping from colours to patterns can also be fine-tuned for optimal results.
CUHK is now looking for potential licensees from around the world to commercialise any of the three manga technologies by incorporating them in commercial software or as plug-ins. The research team is also open to other options such as licensing one or more of the technologies to manga production houses and advertising agencies.
Contact information: laura.yu [at] isis.ox.ac.uk (Laura Yu)
• This article is part of the “ITF Projects Commercialisation Programme” hosted by Hong Kong Science and Technology Parks Corporation (HKSTP) in conjunction with Isis Innovation, the technology transfer company of the University of Oxford. The programme aims to expose Hong Kong innovations to international audiences and hence, facilitate commercialisation via the network of Isis. For information on all projects in the programme, please visit
This project is organised by Hong Kong Science & Technology Parks Corporation. Any opinions, findings, conclusions or recommendations expressed in this material/event (or by members of the project team) do not reflect the views of the Government of Hong Kong Special Administrative Region, the Innovation and Technology Commission or the Vetting Committee of the General Support Programme of the Innovation and Technology Fund.