[2021-12-24] Prof. Po-Chyi Su, National Central University, "Scene Text Detection and Recognition – A Traditional Chinese Scene Text Recognition Competition"

Poster:Post date:2021-12-13
Title: Scene Text Detection and Recognition – A Traditional Chinese Scene Text Recognition Competition
Date: 2021-12-24 3:40pm-5:00pm
Location: R103, CSIE
Speaker: Prof. Po-Chyi Su, National Central University
Hosted by: Prof. Chu-Song Chen


Texts appearing in images are often regions of interest. Locating such areas and then recognizing the contents help to extract image-related information and can facilitate many potential applications. Scene text recognition is considered a challenging scenario since texts in pictures captured by typical cameras may have diversified fonts or sizes, be occluded by other objects and be captured from varying angles or under different light conditions. In this talk, we will first review the related techniques of scene text detection and recognition. Next, considering that Traditional Chinese receives less attention in existing contests, we thus held “AICUP2021 Traditional Chinese Scene Text Recognition Competition” and will describe the setup of the competition and some results. Finally, we will present what we have achieved in this field, share some experiences, and point out future work for further discussions.

Dr. Po-Chyi Su received the B.S. degree from the National Taiwan University, Taipei, Taiwan, in 1995 and the M.S. and Ph.D. degrees from the University of Southern California, Los Angeles, in 1998 and 2003, respectively, all in Electrical Engineering. He then joined Industrial Technology Research Institute, HsinChu, Taiwan, as an engineer. Since August 2004, he has been with the Department of Computer Science and Information Engineering, National Central University, Taiwan. He is now a Professor and serves as the Department Chair. His research interests include multimedia signal processing, multimedia compression, and deep learning.
Last modification time:2021-12-13 AM 9:10

cron web_use_log