Facial outlook comparative analysis

Abstract

It is a software project which uses Eigenface to achieve face recognition and facial outlook ranking (beauty score). Eigenface is a method using the concept of Principal component analysis (PCA) which can reduce a complex data set to a lower dimension and extract the important information. The programming language using in the project is python. The project processes can consist of three parts. The first one is to implement the Eigenface algorithm to compare images. Secondly, collecting images with the facial outlook rank for the facial outlook analysis. The final part is the graphical user interface (GUI) of the project. The methods involve this project are Eigenface and K-nearest neighbors and the ready-made classifier. By implementing eigenface, the similarity of the test image and the training set can be found. The K nearest neighbors of the test image will be used to calculate the average ranking to represent the ranking of the test image. In addition, the ready-made classifier can implement the face detection for importing test image. The software has a main window and two sub-windows for different functions. And it will be packed as a .exe file which can run without installing the python libraries of the program.