Optical-Flow-Based-Movement-Detection

Overview

This project implements a computer vision technique to detect motion between consecutive frames using dense optical flow.

Key Features

• Preprocesses images by converting them to grayscale and applying Gaussian blur.
• Computes dense optical flow using the Farneback algorithm.
• Generates and visualizes a motion mask based on optical flow magnitude.
• Detects contours and filters bounding boxes around moving regions.
• Applies non-maximum suppression to refine bounding boxes.
• Visualizes the detected movements on the original frames.

Dataset

• The dataset used here is M-30-HD (1.7GB) (1200x720)
• We can also use M-30 (290MB) (800x480)
• M-30-HD M-30