Performance evaluation method for mobile computer vision systems using augmented reality

Jonas Nilsson, Anders C.E. Ödblom, Jonas Fredriksson, Adeel Zafar, Fahim Ahmed

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

5 Citations (Scopus)

Abstract

This paper describes a framework which uses augmented reality for evaluating the performance of mobile computer vision systems. Computer vision systems use primarily image data to interpret the surrounding world, e.g to detect, classify and track objects. The performance of mobile computer vision systems acting in unknown environments is inherently difficult to evaluate since, often, obtaining ground truth data is problematic. The proposed novel framework exploits the possibility to add virtual agents into a real data sequence collected in an unknown environment, thus making it possible to efficiently create augmented data sequences, including ground truth, to be used for performance evaluation. Varying the content in the data sequence by adding different virtual agents is straightforward, making the proposed framework very flexible. The method has been implemented and tested on a pedestrian detection system used for automotive collision avoidance. Preliminary results show that the method has potential to replace and complement physical testing, for instance by creating collision scenarios, which are difficult to test in reality.
LanguageEnglish
Title of host publication2010 IEEE Virtual Reality Conference (VR)
Place of PublicationPiscataway, NJ
PublisherIEEE
Pages19-22
Number of pages4
ISBN (Electronic)9781424462377
DOIs
Publication statusPublished - 8 Apr 2010

Fingerprint

Augmented reality
Computer vision
Collision avoidance
Testing

Keywords

  • mobile computing
  • augmented reality
  • virtual reality

Cite this

Nilsson, J., Ödblom, A. C. E., Fredriksson, J., Zafar, A., & Ahmed, F. (2010). Performance evaluation method for mobile computer vision systems using augmented reality. In 2010 IEEE Virtual Reality Conference (VR) (pp. 19-22). Piscataway, NJ: IEEE. https://doi.org/10.1109/VR.2010.5444821
Nilsson, Jonas ; Ödblom, Anders C.E. ; Fredriksson, Jonas ; Zafar, Adeel ; Ahmed, Fahim. / Performance evaluation method for mobile computer vision systems using augmented reality. 2010 IEEE Virtual Reality Conference (VR). Piscataway, NJ : IEEE, 2010. pp. 19-22
@inproceedings{efbcfa98e46c4d49b47f52935831b4af,
title = "Performance evaluation method for mobile computer vision systems using augmented reality",
abstract = "This paper describes a framework which uses augmented reality for evaluating the performance of mobile computer vision systems. Computer vision systems use primarily image data to interpret the surrounding world, e.g to detect, classify and track objects. The performance of mobile computer vision systems acting in unknown environments is inherently difficult to evaluate since, often, obtaining ground truth data is problematic. The proposed novel framework exploits the possibility to add virtual agents into a real data sequence collected in an unknown environment, thus making it possible to efficiently create augmented data sequences, including ground truth, to be used for performance evaluation. Varying the content in the data sequence by adding different virtual agents is straightforward, making the proposed framework very flexible. The method has been implemented and tested on a pedestrian detection system used for automotive collision avoidance. Preliminary results show that the method has potential to replace and complement physical testing, for instance by creating collision scenarios, which are difficult to test in reality.",
keywords = "mobile computing, augmented reality, virtual reality",
author = "Jonas Nilsson and {\"O}dblom, {Anders C.E.} and Jonas Fredriksson and Adeel Zafar and Fahim Ahmed",
year = "2010",
month = "4",
day = "8",
doi = "10.1109/VR.2010.5444821",
language = "English",
pages = "19--22",
booktitle = "2010 IEEE Virtual Reality Conference (VR)",
publisher = "IEEE",

}

Nilsson, J, Ödblom, ACE, Fredriksson, J, Zafar, A & Ahmed, F 2010, Performance evaluation method for mobile computer vision systems using augmented reality. in 2010 IEEE Virtual Reality Conference (VR). IEEE, Piscataway, NJ, pp. 19-22. https://doi.org/10.1109/VR.2010.5444821

Performance evaluation method for mobile computer vision systems using augmented reality. / Nilsson, Jonas; Ödblom, Anders C.E.; Fredriksson, Jonas ; Zafar, Adeel ; Ahmed, Fahim.

2010 IEEE Virtual Reality Conference (VR). Piscataway, NJ : IEEE, 2010. p. 19-22.

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

TY - GEN

T1 - Performance evaluation method for mobile computer vision systems using augmented reality

AU - Nilsson, Jonas

AU - Ödblom, Anders C.E.

AU - Fredriksson, Jonas

AU - Zafar, Adeel

AU - Ahmed, Fahim

PY - 2010/4/8

Y1 - 2010/4/8

N2 - This paper describes a framework which uses augmented reality for evaluating the performance of mobile computer vision systems. Computer vision systems use primarily image data to interpret the surrounding world, e.g to detect, classify and track objects. The performance of mobile computer vision systems acting in unknown environments is inherently difficult to evaluate since, often, obtaining ground truth data is problematic. The proposed novel framework exploits the possibility to add virtual agents into a real data sequence collected in an unknown environment, thus making it possible to efficiently create augmented data sequences, including ground truth, to be used for performance evaluation. Varying the content in the data sequence by adding different virtual agents is straightforward, making the proposed framework very flexible. The method has been implemented and tested on a pedestrian detection system used for automotive collision avoidance. Preliminary results show that the method has potential to replace and complement physical testing, for instance by creating collision scenarios, which are difficult to test in reality.

AB - This paper describes a framework which uses augmented reality for evaluating the performance of mobile computer vision systems. Computer vision systems use primarily image data to interpret the surrounding world, e.g to detect, classify and track objects. The performance of mobile computer vision systems acting in unknown environments is inherently difficult to evaluate since, often, obtaining ground truth data is problematic. The proposed novel framework exploits the possibility to add virtual agents into a real data sequence collected in an unknown environment, thus making it possible to efficiently create augmented data sequences, including ground truth, to be used for performance evaluation. Varying the content in the data sequence by adding different virtual agents is straightforward, making the proposed framework very flexible. The method has been implemented and tested on a pedestrian detection system used for automotive collision avoidance. Preliminary results show that the method has potential to replace and complement physical testing, for instance by creating collision scenarios, which are difficult to test in reality.

KW - mobile computing

KW - augmented reality

KW - virtual reality

U2 - 10.1109/VR.2010.5444821

DO - 10.1109/VR.2010.5444821

M3 - Conference contribution book

SP - 19

EP - 22

BT - 2010 IEEE Virtual Reality Conference (VR)

PB - IEEE

CY - Piscataway, NJ

ER -

Nilsson J, Ödblom ACE, Fredriksson J, Zafar A, Ahmed F. Performance evaluation method for mobile computer vision systems using augmented reality. In 2010 IEEE Virtual Reality Conference (VR). Piscataway, NJ: IEEE. 2010. p. 19-22 https://doi.org/10.1109/VR.2010.5444821