Abstract
This paper presents a concept-based learning framework for sign language recognition. The recognition is based on the concept learning framework which transforms learning about high-level concepts into learning for low-level concepts, the structure of which is simpler and can be learned with fewer samples. The sign language samples were first segmented by a modified Time Varying Parameter (TVP) algorithm which segments the motion time series according to pause points, and the segments are used to build a primitive library with a selforganizing feature map (SOFM) network. Then the SL samples represented by the primitive library were recognized using template matching and model generation methods. Experimental results show that under the framework of concept learning, we achieve 3% to 4% accuracy improvement for the recognition of 95 classes Australian Sign Language (AUSL) using Dynamic Time Warping (DTW) and Hidden Markov Model (HMM) classifiers.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
