Skip to content

lcdurann/Correlation-Techniques-for-Face-Recognition

 
 

Repository files navigation

Correlation Filters for Pattern Recognition

In this Repo you will find documentation of the process of exploring correlation filters and Fourier transforms to identify a human face using advanced correlation techniques. It is a project developed for the course on Optics and Acoustics from the National University of Colombia. It is carried out by:

  • Diego Alejandro Herrera rojas
  • Luis Neme
  • Andrés Duque
  • Juan Manuel Moreno

Description

The aim is to develop a MATLAB - based application that may be deployed on any computer, that ultimately will identify a persons face in a scene. If possible, an Android application might be developed so that the face-recognition system can be used in a smartphone.

Contents

The project contains the following folders:

  1. MATLABCodes: This folder contains the MATLAB codes used to explore the capabilities of correlation techniques.
  2. images: Contains images for filter synthesis and further facial identification processes.
  3. Figures: Contains images for initial exploration.

Push Formats

In order to keep track of the work, the following formats are proposed for committing files to the repository:

  • All code files must me in a directory with the name (Language)Codes. For instance, if the codes are in MATLAB, they must be in the directory MATLABCodes.

  • All code files must have a heading that includes: 1) Descriptive title, 2) Author's name, 3) Date of deploy, 4) Brief description. Codes must be accurately commented so that anyone can understand them.

  • Al images that are not used for filter synthesis must go in the directory Figures. Images that are used for filter synthesis must be in directory images, and in a subdirectory with the name of the person to be identified.

  • Al images must be stored in PNG format.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • MATLAB 96.2%
  • Python 3.8%