Towards open-set text recognition via label-to-prototype learning

Chang Liu, Chun Yang, Hai-Bo Qin, Xiaobin Zhu, Cheng-Lin Liu, Xu-Cheng Yin

Abstract: Highlights•Formulating a new open-set task requires spotting and cognizing novel characters.•Proposing a new framework that handles novel characters without retraining.•Proposing a fast rectification technique for text recognition.

External IDs:doi:10.1016/j.patcog.2022.109109