Academic & Independent Research

Scientific Contributions &
Publications.

Exploring the intersections of machine learning, neural network optimization, and computer vision.

Conference Paper
Published: 2024
Springer Nature

Optimizing American Sign Language Recognition with Binarized Neural Networks: A Comparative Study with Traditional Models

Abstract

Sign language is crucial for communication among individuals with hearing or speech impairments. Automated recognition systems are essential for learning and translating different sign language variants. However, these systems often face high computational demands.

This study proposes a Binarized Neural Network (BNN) architecture designed to drastically reduce memory footprint and computational requirements by restricting weights and activations to +1 and -1. We conduct a comprehensive comparative analysis between our proposed BNN and traditional Convolutional Neural Networks (CNNs) for American Sign Language (ASL) recognition. Results demonstrate that BNNs can achieve competitive accuracy while offering significant advantages in deployment on edge devices and low-power hardware, paving the way for more accessible and real-time sign language translation technologies.

DOI: 10.1007/978-3-032-11335-1_27
View on Springer
Next Step

Bridging Research & Real-world Systems

My academic work informs the intelligent systems I build. Let's discuss how deep learning and optimized architecture can drive your next project.

View Technical Portfolio