Object & Person Detection with Live Video Feed

This project implements Real-time Object & Person Detection using TensorFlow.js with the COCO-SSD model. It captures live video using the user's webcam and detects various objects in real-time, displaying them on the video feed. The application also provides speech feedback about the detected objects, offering an interactive experience.

Features

• Live Object Detection: Real-time object recognition through webcam using the COCO-SSD model.
• Speech Feedback: Detection results are spoken aloud.
• Customizable Object Colors: Different colors for various detected objects, including persons, phones, bottles, and more.
• Real-time Updates: Continuous detection and updating of identified objects on the screen.

Setup Instructions

1. Clone the repository or download the index.html file.
2. Open the index.html file in any modern browser that supports:
   • Webcam access
   • Web Speech API
   • TensorFlow.js

Technologies Used

• TensorFlow.js: For real-time object detection.
• COCO-SSD Model: Pre-trained model for detecting common objects.
• Web Speech API: For speech synthesis (announcing detected objects).
• HTML5 & CSS: For structuring and styling the page.

How It Works

1. The application accesses the user's webcam.
2. It uses the COCO-SSD model to detect objects in the live video feed.
3. Detected objects are drawn with bounding boxes and labeled.
4. The application announces the detected objects via the Web Speech API.
5. Detected objects are continuously updated in real-time, and the screen is refreshed every frame.

Browser Support

This application works in modern browsers that support the Web Speech API and TensorFlow.js. Ensure the webcam is enabled and accessible.

License

This project is licensed under the MIT License.

For more information on how TensorFlow.js or the COCO-SSD model works, refer to their official documentation:

• TensorFlow.js
• COCO-SSD Model