| Title | Real-time Sign Language Recognition with CNNs and ONNX |
| Research Area | Computer Science and Engineering |
| Abstract | This research project aims to enhance American
Sign Language (ASL) communication using Convolutional
Neural Networks (CNNs) and the Open Neural Network
Exchange (ONNX). ASL is crucial for the Deaf and Hard of
Hearing community. By leveraging the Sign Language
MNIST dataset and a custom CNN architecture, our model
achieves high accuracy in recognizing ASL gestures.
Additionally, we explore the role of ONNX in model export
and real-time inference, ensuring cross-platform
compatibility. Through real-time video analysis, we
demonstrate the effectiveness of our model in capturing ASL
gestures, thereby improving communication. This project not
only advances ASL recognition but also underscores the
potential of deep learning and ONNX in developing practical
communication solutions |
| Keywords | Sign Language Recognition, Convolutional Neural Networks, ONNX Model Export, ASL Gesture Recognition, Deaf Communication, Sign Language MNIST, Real-time Inference, Deep Learning, Accessibility Technology, Crossplatform Compatibility |
| Paper Status | Published |
| Volume | 1 |
| Issue | 1 |
| Published On | 12/03/2025 |
| Published File |
IJSRTD_4825.pdf
|