From Multi-Lingual Character Recognition to Multi-model Biometric Verification and Identification

48 views · Published 6 September 2016 · 1:15:47 · Indexed 4 November 2026

Channel: Microsoft Research · 2016 · Science & Technology

Watch on YouTube

In this talk, I will briefly introduce the research activities in the center of intelligent document image processing (CIDIP),Tsinghua University (Beijing, China). I will share with your our research experience in Multi-Lingual Character Recognition for over 20 years, and our development experience in a series of  THOCR Chinese character (Printed, On-line and Offline handwritten) recognition systems and THOCR Multi-lingual (include Japanese, Korean, Arabic, Tibetan, Uyghur, Kazakh, Kirghiz and Arabic characters etc.) document recognition systems. I will also share with you our recent experience in foraying into Multimodal (including Face, Hand-writer, Signature and Iris) Biometric verification and identification algorithms and systems.

More from this channel