UAM | UAM_Biblioteca | Unified search engine | Scientific Production Portal | UAM Research Data Repository
Biblos-e Archivo
    • español
    • English
  • English 
    • español
    • English
  • Log in
JavaScript is disabled for your browser. Some features of this site may not work without it.

Search Biblos-e Archivo

Advanced Search

Browse

All of Biblos-e ArchivoCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsFacultiesThis CollectionBy Issue DateAuthorsTitlesSubjectsFaculties

My Account

Log inRegister

Statistics

View Usage Statistics

Help

Information about Biblos-e ArchivoI want to submit my workFrequently Asked Questions

UAM_Biblioteca

View Item 
  •   Biblos-e Archivo
  • 2 - Trabajos de estudiantes (tesis doctorales, TFMs, TFGs, etc.)
  • Trabajos de estudiantes (tesis doctorales, TFMs, TFGs, etc.)
  • View Item
  •   Biblos-e Archivo
  • 2 - Trabajos de estudiantes (tesis doctorales, TFMs, TFGs, etc.)
  • Trabajos de estudiantes (tesis doctorales, TFMs, TFGs, etc.)
  • View Item

Continual learning for object detection

Author
Jiménez Muñoz, Carlos
Advisor
García Martín, Álvarountranslated
Entity
UAM. Departamento de Tecnología Electrónica y de las Comunicaciones
Date
2021-09
Subjects
Object Detection; Continual Learning; Deep Learning; Telecomunicaciones
URI
http://hdl.handle.net/10486/700221
Note
Master in Deep Learning for Audio and Video Signal Processing

Licencia de Creative Commons
Esta obra está bajo una licencia de Creative Commons Reconocimiento-NoComercial-SinObraDerivada 4.0 Internacional.

Abstract

This Master Thesis is contextualized in the context of the need to count containersin different cities in Spain for garbage collection. In order to perform this countingautomatically, to save time and resources, it is proposed to use the technique of objectdetection on images captured from a car.At the research proposal level, the hypothesis is as follows: ”it is possible to createa model/system that takes advantage of the knowledge acquired from the cubes of acity to facilitate learning in new cities, without significantly worsening its results inthe previously known city”. To achieve this, one has to deal with a problem deriveddirectly from the training algorithm of a neural network, the Catastrophic Forgetting.This problem implies that, once a model with certain weights is trained to solve a newtask (e.g., recognition of a new class) its results will be worse on the previously knowntask/class.In order to try to solve this problem, a current called Continual Learning (alsoknown as Incremental Learning or Lifelong Learning) arises within the area of Arti-ficial Intelligence, which introduces techniques based on reproduction, regularizationor isolation of parameters, among others, to try to mitigate the effect of CatastrophicForgetting in the results of the model, retaining the knowledge prior to the retrainingof the model with the new data.Specifically, in this project we have worked to achieve a system that is capable ofdetecting containers of new cities without losing accuracy in the detection of previouslyknown cities. To this end, we have carried out annotation tasks, adaptation of a GitHubrepository where the EfficientDet model is used for object detection, measurement andmitigation of catastrophic forgetting and finally creation of an interface to facilitatethe use of the application to a potential user.
Show full item record

Files in this item

Thumbnail
Name
jimenez_muñoz_carlos_tfm.pdf
Size
3.183Mb
Format
PDF

Refworks Export

Google™ Scholar:Jiménez Muñoz, Carlos

This item appears in the following Collection(s)

  • Trabajos de estudiantes (tesis doctorales, TFMs, TFGs, etc.) [19713]

Related items

Showing items related by title, author, creator and subject.

  • Object detection and association in multiview scenarios based on Deep learning 

    Moral de Eusebio, Paula
    2019-07
  • Evaluation of deep learning-based classification and object detection algorithms for event cameras 

    Martín Ameneiro, Francisco Javier
    2021-06
  • Long-Term Stationary Object Detection Based on Spatio-Temporal Change Detection 

    Ortego Hernández, DiegoAutoridad UAM; San Miguel Avedillo, Juan CarlosAutoridad UAM; Martínez Sánchez, José MaríaAutoridad UAM
    2015-12-01
All the documents from Biblos-e Archivo are protected by copyrights. Some rights reserved.
Universidad Autónoma de Madrid. Biblioteca
Contact Us | Send Feedback
We are onFacebookCanal BiblosYouTubeTwitterPinterestWhatsappInstagram

Declaración de accesibilidad

 

 

All the documents from Biblos-e Archivo are protected by copyrights. Some rights reserved.
Universidad Autónoma de Madrid. Biblioteca
Contact Us | Send Feedback
We are onFacebookCanal BiblosYouTubeTwitterPinterestWhatsappInstagram

Declaración de accesibilidad