Deep learning applied to facial recognition

• Grant number: 23/11498-1
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: November 01, 2023
• End date: October 31, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Nina Sumiko Tomita Hirata
• Grantee: Kaique Nunes de Oliveira
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 22/15304-4 - Learning context rich representations for computer vision, AP.TEM

Abstract

Historically, facial recognition by computers has often been seen as something from science fiction. However, in recent years, methods have been developed that are capable of achieving face recognition with very high success rates, making this technology increasingly present in people's daily lives. This great advance was mainly due to the contributions brought by Deep Learning, the availability of vast public databases of photos of human faces, and Convolutional Neural Networks (CNNs). CNNs perform feature extraction from images and learn high-level representations from training data, being widely used, in addition to facial recognition, in various tasks within the field of Computer Vision. This research project aims to develop a facial recognition model, initially using publicly available datasets. In this process, the student will have the opportunity to acquire knowledge in the areas of Computer Vision, Machine Learning and mainly Deep Learning, in addition to the scientific research process. They will also have the chance to develop practical skills for implementing and training CNNs. In the end, the goal is to produce an application to test these models using facial images collected in the university environment.