View-based dynamic object recognition based on human perception

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

21 Citations (Scopus)

Abstract

Psychophysical studies have shown that humans actively exploit temporal information such as contiguity of images in object recognition. We have recently developed a recognition system which uses temporal contiguity to learn extensible representations of objects on-line. The system performs well both on real-world and synthetic data and shows robustness under illumination changes. In this paper, we present results which compare the proposed representation against simple image-based representations of the same complexity using Minkowski Minimum Distance classifiers and Support Vector Machine classifiers. Recognition results for all classifiers show large improvements with incorporated temporal information.

Original languageEnglish
Title of host publicationProceedings - International Conference on Pattern Recognition
Pages768-776
Number of pages9
Volume16
Edition3
Publication statusPublished - 2002
Externally publishedYes

Fingerprint

Object recognition
Classifiers
Support vector machines
Lighting

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Vision and Pattern Recognition
  • Hardware and Architecture

Cite this

Bulthoff, H., Wallraven, C., & Graf, A. (2002). View-based dynamic object recognition based on human perception. In Proceedings - International Conference on Pattern Recognition (3 ed., Vol. 16, pp. 768-776)

View-based dynamic object recognition based on human perception. / Bulthoff, Heinrich; Wallraven, Christian; Graf, Arnulf.

Proceedings - International Conference on Pattern Recognition. Vol. 16 3. ed. 2002. p. 768-776.

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

Bulthoff, H, Wallraven, C & Graf, A 2002, View-based dynamic object recognition based on human perception. in Proceedings - International Conference on Pattern Recognition. 3 edn, vol. 16, pp. 768-776.
Bulthoff H, Wallraven C, Graf A. View-based dynamic object recognition based on human perception. In Proceedings - International Conference on Pattern Recognition. 3 ed. Vol. 16. 2002. p. 768-776
Bulthoff, Heinrich ; Wallraven, Christian ; Graf, Arnulf. / View-based dynamic object recognition based on human perception. Proceedings - International Conference on Pattern Recognition. Vol. 16 3. ed. 2002. pp. 768-776
@inproceedings{2a086e636cd945909c6cc34d4b8abb45,
title = "View-based dynamic object recognition based on human perception",
abstract = "Psychophysical studies have shown that humans actively exploit temporal information such as contiguity of images in object recognition. We have recently developed a recognition system which uses temporal contiguity to learn extensible representations of objects on-line. The system performs well both on real-world and synthetic data and shows robustness under illumination changes. In this paper, we present results which compare the proposed representation against simple image-based representations of the same complexity using Minkowski Minimum Distance classifiers and Support Vector Machine classifiers. Recognition results for all classifiers show large improvements with incorporated temporal information.",
author = "Heinrich Bulthoff and Christian Wallraven and Arnulf Graf",
year = "2002",
language = "English",
volume = "16",
pages = "768--776",
booktitle = "Proceedings - International Conference on Pattern Recognition",
edition = "3",

}

TY - GEN

T1 - View-based dynamic object recognition based on human perception

AU - Bulthoff, Heinrich

AU - Wallraven, Christian

AU - Graf, Arnulf

PY - 2002

Y1 - 2002

N2 - Psychophysical studies have shown that humans actively exploit temporal information such as contiguity of images in object recognition. We have recently developed a recognition system which uses temporal contiguity to learn extensible representations of objects on-line. The system performs well both on real-world and synthetic data and shows robustness under illumination changes. In this paper, we present results which compare the proposed representation against simple image-based representations of the same complexity using Minkowski Minimum Distance classifiers and Support Vector Machine classifiers. Recognition results for all classifiers show large improvements with incorporated temporal information.

AB - Psychophysical studies have shown that humans actively exploit temporal information such as contiguity of images in object recognition. We have recently developed a recognition system which uses temporal contiguity to learn extensible representations of objects on-line. The system performs well both on real-world and synthetic data and shows robustness under illumination changes. In this paper, we present results which compare the proposed representation against simple image-based representations of the same complexity using Minkowski Minimum Distance classifiers and Support Vector Machine classifiers. Recognition results for all classifiers show large improvements with incorporated temporal information.

UR - http://www.scopus.com/inward/record.url?scp=0742318497&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0742318497&partnerID=8YFLogxK

M3 - Conference contribution

VL - 16

SP - 768

EP - 776

BT - Proceedings - International Conference on Pattern Recognition

ER -