Proposal and Experimental Results of a DNN Based Real-Time Recognition Method for Ohsone Style Fingerspelling in Static Characters Environment

In Japan, the Employment Promotion Act for Persons with Disabilities was amended in April 2020. The employers are required to hire person with disabilities. On the other hand, the hearing-impaired person have hard in working due to communication problems with the not hearing-impaired person. In order to solve this problem, the not hearing-impaired person needs to learn the fingerspelling. However, fingerspelling is difficult to learn because of limited learning opportunities. In this paper, we propose a real-time recognition method for learning support the Ohsone style fingerspelling.