Challenges in Realizing Artificial Intelligence Assisted Sign Language Recognition
Publication Name
Communications in Computer and Information Science
Abstract
Many publications tout the ability to develop or address certain aspects of research which tries to solve an existing problem. However, some approaches would appear to solve a certain aspect of a problem, but not the real problem at hand. In the context of a challenging real problem, such approach would simply deceive the researcher into extrapolating the existing capabilities but would not offer any practical realization of the real problem. Computer vision-based sign language recognition is one such problem. With our research into hand gesture recognition of dynamic gestures which can easily be compared to different sign languages present in many parts of the world, certain aspects such as static hand sign recognition has no part to play in dynamic hand gesture recognition or in sign language recognition. This article tries to reach out to the readers and researchers to highlight why sign language recognition is currently not making any progress despite enormous inroads into objects detection in realtime using artificial intelligent tools such as YOLO algorithms.
Open Access Status
This publication is not available as open access
Volume
2014 CCIS
First Page
161
Last Page
168