Object Tracking using OpenCV

This Python script demonstrates object tracking in a video using OpenCV. The script utilizes basic object detection techniques and centroid tracking to monitor and label objects moving within a video feed.

2023-12-22.13-35-46.video-converter.com.mp4

Requirements

• Python 3.x
• OpenCV (cv2)
• NumPy (numpy)

Installation

1. Clone the repository:

   git clone https://github.com/your-username/object-tracking-opencv.git
   cd object-tracking-opencv

2. Install dependencies:

   pip install -r requirements.txt

Usage

1. Place your video file (e.g., "2.mp4") in the project directory.

2. Run the script:

   python object_tracking.py

3. Press the Esc key to exit the application.

Explanation

• The script initializes an object detection class and reads a video file frame by frame.
• It detects objects in each frame, computes their centroids, and tracks their movement.
• Objects are tracked by comparing their positions across consecutive frames.
• Detected objects are labeled with unique IDs and displayed on the video feed.

Code Structure

The code consists of several main sections:

• Initialization: Importing necessary libraries, initializing variables, and setting up the object detection class.
• Object Detection: Detecting objects in each frame and extracting their bounding boxes and centroid positions.
• Object Tracking: Comparing centroid positions between frames to track objects and assigning unique IDs.
• Visualization: Drawing bounding boxes, centroids, and labels on the video frame for tracked objects.

Notes

• Tweak the distance threshold and other parameters for object tracking based on your specific use case.
• This code serves as a basic demonstration and can be extended for various object tracking applications.

Contributions

Contributions are welcome! If you have any suggestions, improvements, or feature additions, feel free to open an issue or create a pull request.