This project implements background subtraction using Gaussian Mixture Models (GMM), a probabilistic approach for segmenting image and video frames into foreground and background components. It is particularly useful in tasks like motion detection, object tracking, and surveillance.
- Background Modeling: Utilizes per-pixel GMM to distinguish between background and foreground in training frames.
- Foreground Detection: Subtracts the background and applies thresholding to accurately isolate and detect foreground objects in test frames.
- Flexibility: Capable of handling dynamic scenes with varying lighting conditions and non-stationary backgrounds.