Paper Contents
Abstract
The main form of communication for those who have hearing loss is sign language. This language mostly uses non-manual gestures and hand articulations. Recognition of sign language has gained popularity recently. In this paper, we propose a trainable deep learning network that can efficiently capture the spatiotemporal information from a limited number of sign frames for isolated sign language detection. Three networksthe dynamic motion network (DMN), the accumulative motion network (AMN), and the sign recognition network (SRN)make up our proposed hierarchical sign learning module. In addition, we suggest a method for addressing the variances in the sign samples produced by various signers by extracting essential postures. These crucial postures let the DMN stream acquire the spatiotemporal details relevant to the symptoms. We also provide a cutting-edge method for encapsulating both static and dynamic information about sign motions in a single frame. The main postures of the sign are fused in the forward and backward directions to produce an accumulative video motion frame, preserving the sign's spatial and temporal information. The retrieved features from this frame were combined with the DMN features to be supplied into the SRN for the learning and categorization of signs. This frame was used as input to the AMN stream. The suggested method is effective for recognizing solitary sign language, particularly for static signs.
Copyright
Copyright © 2025 Rushikesh Takik. This is an open access article distributed under the Creative Commons Attribution License.
