Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot

Convolutional Neural Networks are usually fitted with manually labelled data. The labelling process is very time-consuming since large datasets are required. The use of external hardware may help in some cases, but it also introduces noise to the labelled data. In this paper, we pose a new data labe...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Claudia Álvarez-Aparicio, Ángel Manuel Guerrero-Higueras, Luis V. Calderita, Francisco J. Rodríguez-Lera, Vicente Matellán, Camino Fernández-Llamas
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	bootstrapping convolutional neural networks LIDAR PeTra re-training robotics Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
Acceso en línea:	https://doaj.org/article/a73eebb439e341b69b5b95c61ec8ebfc
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:a73eebb439e341b69b5b95c61ec8ebfc
record_format	dspace
spelling	oai:doaj.org-article:a73eebb439e341b69b5b95c61ec8ebfc2021-11-11T15:07:22ZConvolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot10.3390/app1121100432076-3417https://doaj.org/article/a73eebb439e341b69b5b95c61ec8ebfc2021-10-01T00:00:00Zhttps://www.mdpi.com/2076-3417/11/21/10043https://doaj.org/toc/2076-3417Convolutional Neural Networks are usually fitted with manually labelled data. The labelling process is very time-consuming since large datasets are required. The use of external hardware may help in some cases, but it also introduces noise to the labelled data. In this paper, we pose a new data labelling approach by using bootstrapping to increase the accuracy of the PeTra tool. PeTra allows a mobile robot to estimate people’s location in its environment by using a LIDAR sensor and a Convolutional Neural Network. PeTra has some limitations in specific situations, such as scenarios where there are not any people. We propose to use the actual PeTra release to label the LIDAR data used to fit the Convolutional Neural Network. We have evaluated the resulting system by comparing it with the previous one—where LIDAR data were labelled with a Real Time Location System. The new release increases the <i>MCC</i>-score by 65.97%.Claudia Álvarez-AparicioÁngel Manuel Guerrero-HiguerasLuis V. CalderitaFrancisco J. Rodríguez-LeraVicente MatellánCamino Fernández-LlamasMDPI AGarticlebootstrappingconvolutional neural networksLIDARPeTrare-trainingroboticsTechnologyTEngineering (General). Civil engineering (General)TA1-2040Biology (General)QH301-705.5PhysicsQC1-999ChemistryQD1-999ENApplied Sciences, Vol 11, Iss 10043, p 10043 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	bootstrapping convolutional neural networks LIDAR PeTra re-training robotics Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
spellingShingle	bootstrapping convolutional neural networks LIDAR PeTra re-training robotics Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999 Claudia Álvarez-Aparicio Ángel Manuel Guerrero-Higueras Luis V. Calderita Francisco J. Rodríguez-Lera Vicente Matellán Camino Fernández-Llamas Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot
description	Convolutional Neural Networks are usually fitted with manually labelled data. The labelling process is very time-consuming since large datasets are required. The use of external hardware may help in some cases, but it also introduces noise to the labelled data. In this paper, we pose a new data labelling approach by using bootstrapping to increase the accuracy of the PeTra tool. PeTra allows a mobile robot to estimate people’s location in its environment by using a LIDAR sensor and a Convolutional Neural Network. PeTra has some limitations in specific situations, such as scenarios where there are not any people. We propose to use the actual PeTra release to label the LIDAR data used to fit the Convolutional Neural Network. We have evaluated the resulting system by comparing it with the previous one—where LIDAR data were labelled with a Real Time Location System. The new release increases the <i>MCC</i>-score by 65.97%.
format	article
author	Claudia Álvarez-Aparicio Ángel Manuel Guerrero-Higueras Luis V. Calderita Francisco J. Rodríguez-Lera Vicente Matellán Camino Fernández-Llamas
author_facet	Claudia Álvarez-Aparicio Ángel Manuel Guerrero-Higueras Luis V. Calderita Francisco J. Rodríguez-Lera Vicente Matellán Camino Fernández-Llamas
author_sort	Claudia Álvarez-Aparicio
title	Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot
title_short	Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot
title_full	Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot
title_fullStr	Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot
title_full_unstemmed	Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot
title_sort	convolutional neural networks refitting by bootstrapping for tracking people in a mobile robot
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/a73eebb439e341b69b5b95c61ec8ebfc
work_keys_str_mv	AT claudiaalvarezaparicio convolutionalneuralnetworksrefittingbybootstrappingfortrackingpeopleinamobilerobot AT angelmanuelguerrerohigueras convolutionalneuralnetworksrefittingbybootstrappingfortrackingpeopleinamobilerobot AT luisvcalderita convolutionalneuralnetworksrefittingbybootstrappingfortrackingpeopleinamobilerobot AT franciscojrodriguezlera convolutionalneuralnetworksrefittingbybootstrappingfortrackingpeopleinamobilerobot AT vicentematellan convolutionalneuralnetworksrefittingbybootstrappingfortrackingpeopleinamobilerobot AT caminofernandezllamas convolutionalneuralnetworksrefittingbybootstrappingfortrackingpeopleinamobilerobot
_version_	1718437162437836800

Convolutional Neural Networks Refitting by Bootstrapping for Tracking People in a Mobile Robot

Ejemplares similares