Real Time Sign Language Detection

Inspiration

Imagine a world where communication isn't limited by the ability to hear. With the help of technology, we can bridge the gap between the deaf community and the hearing community. Our hand sign recognition project is a step towards this goal.

This project is not only about technology, it's about inclusion and breaking down barriers. It's about empowering individuals to express themselves and be understood. We are proud to be a part of this effort and look forward to seeing the positive impact it will have on the deaf community."

What it does

This project uses machine learning and TensorFlow to recognize human hand signs, providing a new form of communication for deaf individuals. The model is trained on a dataset of hand sign images and can classify new images of hand signs with over 90% accuracy.

How we built it

• Collected and annotated a dataset of hand-sign images using a combination of online resources and personal photographs.
• Pre-processed the photos to ensure a uniform size and lighting conditions.
• Used a pre-trained computer vision as a starting point and fine-tuned it on our dataset using TensorFlow.
• Tested the model's performance on a separate validation dataset and iteratively made adjustments to improve accuracy.

Challenges we ran into

• Obtaining a large and diverse dataset of hand sign images for training the model.
• Fine-tuning the model to achieve high accuracy while also reducing the number of false positives.
• Ensuring the model's performance on a wide range of hand signs and variations

Accomplishments that we're proud of

• Developing a machine learning model that can recognize human hand signs with over 90% accuracy.
• Contributing to the effort of creating inclusive and accessible communication tools for the deaf community.

What we learned

• The importance of collecting a diverse and representative dataset for training machine learning models.
• The power of transfer learning in reducing the amount of data and computational resources required for training a new model.
• The potential for machine learning to help bridge communication gaps and improve accessibility.

What's next for Real Time Sign Language Detection

• Expanding the model's capabilities to recognize more hand signs and variations.
• Improving the model's performance on low-quality or real-world images.
• Integrating the model into a mobile or web application to make it more widely accessible.
• Testing the model with deaf individuals and incorporating their feedback to improve the model's usability.

Built With

• opencv
• python
• tensorflow