Final Project for COGS 118B
In this project, our group decided to use FaceTracer Database, which is a large collectionof real-world face images, collected from the internet from the University of Columbia in orderto construct a support vector machine (SVM) and neural network which would correctly classifythe age of a person in a given picture. Initially starting with a different data set, our group’scuriosity was to see the role that facial features played when trying to categorize importantfeatures such as age or age group. We found that the FaceTracer Database was the most precisedataset and had all of the information we were looking for, such as a clear age group, facialangles, and relationships between facial features. Face classification can be an especially usefultool for social media companies or anyone looking to detect certain aspects of a person throughtheir face– aspects such as age, emotion or facial hair. Some uses of facial classification arepreventing crimes like shoplifting or cheating in casinos, security in unlocking phones, andtargeted advertising. Its potential societal uses drove our group to learn more about whatalgorithms gave the best results and better understand the implementation behind thosealgorithms.