HackEd-2024 Project

Participants:

Aditya Patel
Siddhant Goel
Kashish Gupta

NOTE:

This Github Repo doesn't contain the fully trained data model. So, the fully trained files with ML model can be accessed at the following Google Drive Link: HackEd 2024 - Google Drive(with Trained Data Files)

PROBLEM STATEMENT:

We have created Real-Time Sign Language Detection using Tensorflow Object Detection and Python | Deep Learning SSD*

In this project, our collective goal was to develop a real-time sign language detection system using deep learning techniques with Tensorflow and Python. Implementing a Single Shot Multibox Detector (SSD) for object detection, specifically tailored to recognize and interpret sign language gestures was our primary focus in the project.

DESCRIPTION:

The primary objective of this project was to create a robust and efficient system capable of recognizing and interpreting sign language gestures in real-time using laptop camera. The implementation revolves around leveraging the power of Tensorflow's Object Detection API and utilizing the SSD architecture for accurate and fast detection of sign language gestures.

KEY COMPONENTS:

1. Tensorflow Object Detection API: This project heavily relies on Tensorflow's Object Detection API to facilitate the implementation of an accurate and efficient detection model. This API offers a rich set of pre-trained models and tools for custom model training.

2. Single Shot Multibox Detector (SSD): SSD is a state-of-the-art object detection algorithm known for its real-time processing capabilities. By utilizing SSD, the project aims to achieve high accuracy in recognizing and localizing sign language gestures within video streams.

WORKFLOW:

1. Data Collection: Gather a diverse dataset of sign language gestures, ensuring it covers a wide range of signs. Each sign is associated with a unique label.

2. Model Training: Utilize Tensorflow Object Detection API to train the SSD model on the collected sign language dataset. Fine-tune the pre-trained model to adapt to the specific nuances of sign language gestures.

3. Real-Time Detection: Implement a real-time sign language detection system using the trained SSD model. This involves capturing video frames, processing them through the model, and identifying sign language gestures in the live stream.

4. Integration with User Interface: Develop a user-friendly interface that visually displays the real-time sign language detection results. The interface should provide an intuitive and accessible way for users to interact with the system.