The scripts used for calligraphic works are roughly divided into five scripts, Seal, Clerical, Cursive, Running, and Standard. And many of these require specialized knowledge for reading. This method provides image recognition and transformation to support the appreciation of calligraphic works. I treated the former as my thesis and now I am working on the latter.
|Shohei Ninomiya||Keio University|
|Masanori Nakayama||Keio University|
|Atsushi Miyazawa||Keio University|
demo video (×2)
- Shohei Ninomiya, Masanori Nakayama, Atsushi Miyazawa, Issei Fujishiro: “Toward script translation in calligraphic works using deep learning: Character recognition of seal scripts,” in The Technical Report of the Institute of Image Information and Television Engineers, Vol. 44, No. 10, pp. 75―78, March, 2020, Excellent Research Presentation Award (in Japanese).
- Shohei Ninomiya, Masanori Nakayama, Atsushi Miyazawa, Issei Fujishiro: “Script translation in calligraphic works using deep learning,” in Proceedings of the 82th National Convention of International Processing Society of Japan, Vol. 4, pp. 135―136 (2ZC-04), Kanazawa University Ohgigaoka Campus, Ishikawa, Online, March 5―7, 2020, Student Encouragement Award (in Japanese).
- Grant-in-Aid for Scientific Research (A): 17H00737 (2019―)