- Juan Camilo Ceballos Arias
- Miguel Angel Mejia Muñoz
- Juan Esteban Torres Marulanda
- Danny Styvens Cardona Pineda
The aim of the project is to implement methodologies or if possible to propose improvements to identify human faces. To perform this task, Eigenfaces method will be used together with a previous preparation of the images.
Initially, outlier images are identified for face and landscape datasets, using metrics such as Manhattan, Euclidean, Chebyshev and Minkowsky distances (p = 5/2 and √2/2). Then the obtained results are compared to determine which one of the metrics has better performance in the identification of outliers images.
To replicate this work you must clone the repository in your local machine, in the repository you will find the code, the data and a Notebook from which all the functions are run. Make sure you have Python and Jupyter installed on your computer.
The project is being written in python 3.7, it is recommended to use a version of python superior to 3.5 since packages are used that could present conflicts, for example the cv2 package that is used to manipulate images. The code has been tested on Windows and macOS.
We use Python 3.7