Depth Map Visualisation

A real-time computer vision application that transforms ordinary webcam footage into interactive depth maps using the MiDaS neural network. The system estimates distances of objects from the camera and visualizes them through intuitive color coding - with warm colors indicating nearby objects and cool colors showing distant ones. Features include switchable indoor and outdoor modes, real-time distance measurements, and an interactive color legend. Built with Python, OpenCV, and PyTorch, this project demonstrates practical applications of monocular depth estimation for robotics, AR prototyping, and computer vision research.