Straightforward Recognition of Daily Objects in Smart Environments from Wearable Vision Sensor

Javier Medina Quero, Federico Cruciani, Lorenzo Seidenari, Macarena Espinilla, CD Nugent

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this work, we propose a method to create and
synthesize a new set of virtual images of daily objects within
a smart environment partially automating the labeling process.
Proposed methods enable the generation of a large dataset
from a set of few images using an ad hoc data augmentation,
which increases the original dataset size, generating new items
through partial modification of available images. The proposed
method for data augmentation is accomplished through the
following steps: (i) object tracking is proposed to identify and
label static objects; and (ii) background subtraction is used to
select the masked foreground object of moving objects, which
are virtually projected with geometry transformation over room
images used as background. Furthermore, a case study is carried
out, where an inhabitant wears a wearable vision sensor in a daily
scene. Eight objects are learned using the proposed methodology.
Finally, obtained results and successful recognition rates are
discussed.
LanguageEnglish
Title of host publication2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019
Pages46-51
Number of pages6
ISBN (Electronic)9781538691519
DOIs
Publication statusPublished - 6 Jun 2019
Event2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019: PerCom Pervasive Computing 2019 - Kyoto, Japan
Duration: 11 Mar 201915 Mar 2019
http://www.percom.org/Previous/ST2019/home.html

Conference

Conference2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019
Abbreviated titlePerCom 2019
CountryJapan
CityKyoto
Period11/03/1915/03/19
Internet address

Fingerprint

Labeling
Wear of materials
Geometry
Sensors

Keywords

  • activity recognition
  • data augmentation
  • deep learning
  • object recognition
  • smart environments
  • wearable vision sensors

Cite this

Quero, J. M., Cruciani, F., Seidenari, L., Espinilla, M., & Nugent, CD. (2019). Straightforward Recognition of Daily Objects in Smart Environments from Wearable Vision Sensor. In 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019 (pp. 46-51). [8730860] https://doi.org/10.1109/PERCOMW.2019.8730860
Quero, Javier Medina ; Cruciani, Federico ; Seidenari, Lorenzo ; Espinilla, Macarena ; Nugent, CD. / Straightforward Recognition of Daily Objects in Smart Environments from Wearable Vision Sensor. 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019. 2019. pp. 46-51
@inproceedings{31f259200b2441bd8d86f8202150a825,
title = "Straightforward Recognition of Daily Objects in Smart Environments from Wearable Vision Sensor",
abstract = "In this work, we propose a method to create andsynthesize a new set of virtual images of daily objects withina smart environment partially automating the labeling process.Proposed methods enable the generation of a large datasetfrom a set of few images using an ad hoc data augmentation,which increases the original dataset size, generating new itemsthrough partial modification of available images. The proposedmethod for data augmentation is accomplished through thefollowing steps: (i) object tracking is proposed to identify andlabel static objects; and (ii) background subtraction is used toselect the masked foreground object of moving objects, whichare virtually projected with geometry transformation over roomimages used as background. Furthermore, a case study is carriedout, where an inhabitant wears a wearable vision sensor in a dailyscene. Eight objects are learned using the proposed methodology.Finally, obtained results and successful recognition rates arediscussed.",
keywords = "activity recognition, data augmentation, deep learning, object recognition, smart environments, wearable vision sensors",
author = "Quero, {Javier Medina} and Federico Cruciani and Lorenzo Seidenari and Macarena Espinilla and CD Nugent",
note = "REF Non Compliant but it does not have an ISSN so does not need to comply",
year = "2019",
month = "6",
day = "6",
doi = "10.1109/PERCOMW.2019.8730860",
language = "English",
pages = "46--51",
booktitle = "2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019",

}

Quero, JM, Cruciani, F, Seidenari, L, Espinilla, M & Nugent, CD 2019, Straightforward Recognition of Daily Objects in Smart Environments from Wearable Vision Sensor. in 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019., 8730860, pp. 46-51, 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019, Kyoto, Japan, 11/03/19. https://doi.org/10.1109/PERCOMW.2019.8730860

Straightforward Recognition of Daily Objects in Smart Environments from Wearable Vision Sensor. / Quero, Javier Medina; Cruciani, Federico; Seidenari, Lorenzo; Espinilla, Macarena; Nugent, CD.

2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019. 2019. p. 46-51 8730860.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Straightforward Recognition of Daily Objects in Smart Environments from Wearable Vision Sensor

AU - Quero, Javier Medina

AU - Cruciani, Federico

AU - Seidenari, Lorenzo

AU - Espinilla, Macarena

AU - Nugent, CD

N1 - REF Non Compliant but it does not have an ISSN so does not need to comply

PY - 2019/6/6

Y1 - 2019/6/6

N2 - In this work, we propose a method to create andsynthesize a new set of virtual images of daily objects withina smart environment partially automating the labeling process.Proposed methods enable the generation of a large datasetfrom a set of few images using an ad hoc data augmentation,which increases the original dataset size, generating new itemsthrough partial modification of available images. The proposedmethod for data augmentation is accomplished through thefollowing steps: (i) object tracking is proposed to identify andlabel static objects; and (ii) background subtraction is used toselect the masked foreground object of moving objects, whichare virtually projected with geometry transformation over roomimages used as background. Furthermore, a case study is carriedout, where an inhabitant wears a wearable vision sensor in a dailyscene. Eight objects are learned using the proposed methodology.Finally, obtained results and successful recognition rates arediscussed.

AB - In this work, we propose a method to create andsynthesize a new set of virtual images of daily objects withina smart environment partially automating the labeling process.Proposed methods enable the generation of a large datasetfrom a set of few images using an ad hoc data augmentation,which increases the original dataset size, generating new itemsthrough partial modification of available images. The proposedmethod for data augmentation is accomplished through thefollowing steps: (i) object tracking is proposed to identify andlabel static objects; and (ii) background subtraction is used toselect the masked foreground object of moving objects, whichare virtually projected with geometry transformation over roomimages used as background. Furthermore, a case study is carriedout, where an inhabitant wears a wearable vision sensor in a dailyscene. Eight objects are learned using the proposed methodology.Finally, obtained results and successful recognition rates arediscussed.

KW - activity recognition

KW - data augmentation

KW - deep learning

KW - object recognition

KW - smart environments

KW - wearable vision sensors

UR - http://www.scopus.com/inward/record.url?scp=85067967374&partnerID=8YFLogxK

U2 - 10.1109/PERCOMW.2019.8730860

DO - 10.1109/PERCOMW.2019.8730860

M3 - Conference contribution

SP - 46

EP - 51

BT - 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019

ER -

Quero JM, Cruciani F, Seidenari L, Espinilla M, Nugent CD. Straightforward Recognition of Daily Objects in Smart Environments from Wearable Vision Sensor. In 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019. 2019. p. 46-51. 8730860 https://doi.org/10.1109/PERCOMW.2019.8730860