Real-time camera operation and tracking for the streaming of teaching activities

Master Universitario en Deep Learning for Audio and Video Signal Processing

Detalhes bibliográficos
Autor: Vinuesa Solana, Javier
Formato: tesis de maestría
Fecha de publicación:2021
País:España
Recursos:Universidad Autónoma de Madrid
Repositorio:Biblos-e Archivo. Repositorio Institucional de la UAM
Idioma:inglés
OAI Identifier:oai:repositorio.uam.es:10486/697504
Acesso em linha:http://hdl.handle.net/10486/697504
Access Level:acceso abierto
Palavra-chave:Deep Learning
Neural networks
Convolutional networks
Telecomunicaciones
id ES_0fff11e77f0f2d7703da773f2d2ae8a1
oai_identifier_str oai:repositorio.uam.es:10486/697504
network_acronym_str ES
network_name_str España
repository_id_str
spelling Real-time camera operation and tracking for the streaming of teaching activitiesVinuesa Solana, JavierDeep LearningNeural networksConvolutional networksTelecomunicacionesMaster Universitario en Deep Learning for Audio and Video Signal ProcessingThe primary driving force of this work comes from the Lab’s urgent needs to offer students the opportunity to attend a remote event from home or anywhere in the world in real-time. The main objective of this work is to build a real-time tracker to follow the movements of the lecturer. After that we will build a framework to handle a PTZ (Pan Tilt and Zoom) camera based on the lecturer movements. That is, if the lecturer goes to the left, the camera will turn to the left. To tackle this project we will follow a project developed by Gebrehiwot, A. which involved building a real-time tracker. The problem of this tracker is that was implemented on Ubuntu and running with a very complex CNN which required the use a good GPU on our computer. As Gebrehiwot, A. rightly points out at the end of his report, not everyone has an Ubuntu partition or a GPU on their computers so we started moving the real time tracker to Windows. To achieve this objective we used Anaconda Windows which made our work much easier. After that we implemented a lightweight backbone of the tracker allowing us to run it on computers with a fewer processing power. Once that all this process was done, we put into practice the mentioned framework for handling the movement of the PTZ camera. This framework uses the implemented lightweight tracker to follow the lecturer moves and depending on these movements the camera will pan and tilt automatically. We tested this framework on streaming platforms like YouTube proving that can greatly improve the quality of online classes. Finally we draw conclusions from the work done and propose future work to improve the framework.Bescos Cano, JesúsDepartamento de Tecnología Electrónica y de las ComunicacionesEscuela Politécnica Superior20212021-06-01master thesishttp://purl.org/coar/resource_type/c_bdccNAhttp://purl.org/coar/version/c_be7fb7dd8ff6fe43info:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10486/697504reponame:Biblos-e Archivo. Repositorio Institucional de la UAMinstname:Universidad Autónoma de MadridInglésengopen accesshttp://purl.org/coar/access_right/c_abf2info:eu-repo/semantics/openAccessoai:repositorio.uam.es:10486/6975042026-06-23T12:46:27Z
dc.title.none.fl_str_mv Real-time camera operation and tracking for the streaming of teaching activities
title Real-time camera operation and tracking for the streaming of teaching activities
spellingShingle Real-time camera operation and tracking for the streaming of teaching activities
Vinuesa Solana, Javier
Deep Learning
Neural networks
Convolutional networks
Telecomunicaciones
title_short Real-time camera operation and tracking for the streaming of teaching activities
title_full Real-time camera operation and tracking for the streaming of teaching activities
title_fullStr Real-time camera operation and tracking for the streaming of teaching activities
title_full_unstemmed Real-time camera operation and tracking for the streaming of teaching activities
title_sort Real-time camera operation and tracking for the streaming of teaching activities
dc.creator.none.fl_str_mv Vinuesa Solana, Javier
author Vinuesa Solana, Javier
author_facet Vinuesa Solana, Javier
author_role author
dc.contributor.none.fl_str_mv Bescos Cano, Jesús
Departamento de Tecnología Electrónica y de las Comunicaciones
Escuela Politécnica Superior
dc.subject.none.fl_str_mv Deep Learning
Neural networks
Convolutional networks
Telecomunicaciones
topic Deep Learning
Neural networks
Convolutional networks
Telecomunicaciones
description Master Universitario en Deep Learning for Audio and Video Signal Processing
publishDate 2021
dc.date.none.fl_str_mv 2021
2021-06-01
dc.type.none.fl_str_mv master thesis
http://purl.org/coar/resource_type/c_bdcc
NA
http://purl.org/coar/version/c_be7fb7dd8ff6fe43
dc.type.openaire.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
dc.identifier.none.fl_str_mv http://hdl.handle.net/10486/697504
url http://hdl.handle.net/10486/697504
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Biblos-e Archivo. Repositorio Institucional de la UAM
instname:Universidad Autónoma de Madrid
instname_str Universidad Autónoma de Madrid
reponame_str Biblos-e Archivo. Repositorio Institucional de la UAM
collection Biblos-e Archivo. Repositorio Institucional de la UAM
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869403485073571840
score 15,300724

Registos relacionados