Real-Time Background Replacement using OpenCV

This project performs real-time background subtraction and background replacement using OpenCV and Python.
It detects foreground movement by comparing the current frame with a captured reference background frame and replaces the background with frames from another video source.

Features

• Real-time foreground extraction
• Background replacement using video
• Automatic video looping
• Manual background recapture
• Keyboard controls for interaction
• Works with uploaded video files

---

Technologies Used

• Python
• OpenCV
• NumPy

---

How It Works

1. A reference background frame is captured.
2. Each incoming frame is compared with the reference frame.
3. Differences between frames are extracted.
4. Foreground is separated from the background.
5. The foreground is combined with another background video frame.

---

File Requirements

Upload a video file named:

uploaded_video.h264

Or change the filename in the code:

cv2.VideoCapture('your_video_name.mp4')

---

Installation

Install dependencies:

pip install -r requirements.txt

---

Run the Project

python Augmented_Reality.py

---

Keyboard Controls

Key	Action
q	Capture current frame as background
p	Prepare to capture new background
d	Exit the program

---

Project Structure

.
├── Augmented_Reality.py
├── requirements.txt
├── .gitignore
├── uploaded_video.h264
└── README.md

---

Future Improvements

• Green screen support
• Deep learning segmentation
• Better edge refinement
• Webcam live support
• Multiple background options

---

Example Use Cases

• Virtual background replacement
• Motion detection experiments
• Computer vision learning projects
• Video effects and editing

---

Author

Shubh Sharma