Real-Time Object Detection System using OpenCV and MobileNet SSD

This project implements a real-time object detection system using OpenCV's DNN module and a pre-trained MobileNet SSD model trained on the COCO dataset. It captures video from your webcam and detects 80+ classes of objects with bounding boxes and confidence scores.

📦 Features

• Real-time object detection from webcam input
• Uses pre-trained MobileNet SSD (TensorFlow model)
• Draws bounding boxes and labels with confidence
• Lightweight and fast; suitable for CPU inference

🛠️ Technologies Used

• Python
• OpenCV (cv2)
• TensorFlow SSD (frozen graph)
• COCO Dataset class labels

📁 Files Included

• main.py – Main script for detection
• coco.names – List of class labels from the COCO dataset
• ssd_mobilenet_v3_large_coco_2020_01_14.pbtxt – Configuration file for the model
• frozen_inference_graph.pb – Pre-trained model weights

▶️ Getting Started

1. Clone the Repository

git clone https://github.com/notRamish/object-detection-system.git
cd object-detection-system

⬇️ Install Dependencies

pip install -r requirements.txt

⏯️ Run

python detect.py

To quit; press 'q'