Sixth International Conference on Spoken Language Processing
(ICSLP 2000)

Beijing, China
October 16-20, 2000

A Parallel Multi-Stream Model for Sign Language Recognition

Jiyong Ma, Wen Gao

Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China

In this paper, the sub-units in each stream are used and embedded in the multi-stream model. In this framework, sign language recognition system was implemented and evaluated. Experiments were carried out for 5177 Chinese signs. The real time isolated recognition rate is 95.1%. For continuous sign recognition, the word correct rate is 91.8%. This has shown that parallel multi-stream model is powerful for sign language recognition considering the problem of scalability.


Full Paper

Bibliographic reference.  Ma, Jiyong / Gao, Wen (2000): "A parallel multi-stream model for sign language recognition", In ICSLP-2000, vol.2, 751-754.