This repository contains the works and laboratories of Computer Vision with Deep Learning I do in the Master's degree in Advanced Telecommunication Technologies (MATT), at Universitat Politècnica de Catalunya.
The work in this session will address the problem of finding homographies between images. The homographies will be used in two different aplications: 1) creating panoramas from a set of photographs and 2) augmented reality.
In this practical work we will address the problem of estimating the fundamental matrix between two images. To do this, the characteristic points will be detected in both images and then, correspondences between points will be established. Using these correspondences, the fundamental matrix can be estimated.
This lab will address the issue of finding interesting points and establishing correspondences between them for two similar images (for example, a stereo pair, two images captured to create a panorama, etc.). We are going to detect characteristic points in both images and then establish the correspondences between the detected points. We will use different methods to detect characteristic points:
Harris corner detector SIFT SuperPoints
This practical work has two objectives: first, to analyze the operation of some low-level algorithms. Second, to use these tools to solve practical problems.