2026-04-05T11:33:53Zhttps://recercat.cat/oai/request

oai:recercat.cat:2117/4408142026-01-20T07:40:44Zcom_2072_1033col_2072_452951

Face recognition using deep learning Rodriguez Esteban, Lucas Cristian Pop, Teodor Àrees temàtiques de la UPC::Enginyeria de la telecomunicació Python (Computer program language) Neural networks (Computer science) Artificial intelligence--Engineering applications Deep learning Face recognition Machine learning Neural networks Python Tensorflow Mxnet Opencv Ai Convolutional neural networks Multitask learning Python (Llenguatge de programació) Xarxes neuronals (Informàtica) Intel·ligència artificial--Aplicacions a l'enginyeria The main goal of this project is to obtain the knowledge necessary to understand all the process behind a Face Recognition program. Starting by understanding what is an Artificial Neuron and its different parts, the concept of an Artificial Neural Network, specifically a Convolutional Neural Network and its different layers. This kind of Neural Network is a state-of-art in the field of computer vision. Understanding the concept of Machine learning and Deep learning, including the training and testing steps, which are the fundamental steps in all Deep Learning projects. This project was developed during an exchange program in the National Taiwan University of Science and Technologies, specifically in the department of Computer Science and Information Engineering. The methodology that we followed is the next, every two weeks we did a presentation in front of all members of our lab, and those presentations were divided in two parts, theoretical and practice. The theoretical parts contained a summary of papers and interesting information necessary to develop and understand the practical part. The practical part contained the code in Python and the results obtained, for instance, an image with a face detection. The duration of each presentations was around fifteen minutes plus questions. The final presentation was on 25th June, 2018. We are very proud about the results obtained, first of all because we have obtained the knowledge necessary to understand all the basic theory behind a Face Recognition system, secondly, we have learned a new programming language, Python and most importantly, we have done the training and testing of an Artificial Neural Network. 2018-06-28 Bachelor thesis Open Access Universitat Politècnica de Catalunya