Hierarchical CNN ASL Recognition with Feature Comparison

Abstract

Every day, many individuals live with conditions such as deafness, muteness, or blindness, and they struggle to communicate effectively with others. Effective communication between deaf and hearing individuals is also crucial for accessibility and inclusion in daily life, including education, healthcare, and public services. Sign Language Recognition technology addresses this challenge by providing real-time translation of sign language into text or speech, facilitating smoother and more natural interactions. The increasing importance of SLR lies in its ability to bridge the communication gap fast and accurately, making everyday activities more accessible for deaf people. However, most such approaches to addressing these challenges primarily relied on labeled output from a neural network, but these methods have not provided a comprehensive solution when it comes to immense sign variation.