I took this manuscript photo last time when I visited the Tokyo National Museum (I got the permission). It's not in my whole dataset. I put it in OCR model. The result is not perfect, but good enough to read. It will be like a dream if we can create AR app for real time OCR. pic.twitter.com/ey5jLuAllQ— tkasasagi (@tkasasagi) October 13, 2018
Pages in this blog
Sunday, October 14, 2018
Optical Character Recognition for Japanese manuscript text
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment