Sign language translation based on new continuous sign language dataset
Feng Shi, Tiantian Yuan · 2022 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA) · 2022
deaf individuals rely heavily on one another’s use of sign language as a means of communication. In this era, the significance of accessibility has become more and more important. However, for most people with normal hearing, learning sign language is a very difficult thing. Many research teams around the world are using Deep Learning to develop translators for sign language recognition. However, there is a lack of larger and more semantically rich sign language datasets. In this article, we propose a new Chinese sign language dataset- 109 unique sentences from 50 sign language signers and 27,250 clips. On the new Chinese sign language dataset, we further propose a sequence-to-sequence deep learning approach in order to demonstrate how deep learning may continually lower the communication barriers that exist between persons who are deaf and those who have normal hearing.