A Simulator for Concept Detector Output

Robin Aly, Djoerd Hiemstra

Research output: Book/ReportReportProfessional

14 Downloads (Pure)

Abstract

Concept based video retrieval is a promising search paradigm because it is fully automated and it investigates the fine grained content of a video, which is normally not captured by human annotations. Concepts are captured by so-called concept detectors. However, since these detectors do not yet show a sufficient performance, the evaluation of retrieval systems, which are built on top of the detector output, is difficult. In this report we describe a software package which generates simulated detector output for a specified performance level. Afterwards, this output can be used to execute a search run and ultimately to evaluate the performance of the proposed retrieval method, which is normally done through comparison to a baseline. The probabilistic model of the detectors are two Gaussians, one for the positive and one for the negative class. Thus, the parameters for the simulation are the two means and deviations plus the prior probability of the concept in the dataset.
Original languageUndefined
Place of PublicationEnschede
PublisherCentre for Telematics and Information Technology (CTIT)
Number of pages12
Publication statusPublished - 1 Dec 2009

Publication series

NameCTIT Technical Report Series
PublisherCentre for Telematics and Information Technology, University of Twente
No.TR-CTIT-09-40
ISSN (Print)1381-3625

Keywords

  • Detector Simulation
  • Video Information Retrieval
  • EWI-16544
  • IR-68675
  • METIS-264441

Cite this

Aly, R., & Hiemstra, D. (2009). A Simulator for Concept Detector Output. (CTIT Technical Report Series; No. TR-CTIT-09-40). Enschede: Centre for Telematics and Information Technology (CTIT).
Aly, Robin ; Hiemstra, Djoerd. / A Simulator for Concept Detector Output. Enschede : Centre for Telematics and Information Technology (CTIT), 2009. 12 p. (CTIT Technical Report Series; TR-CTIT-09-40).
@book{feaa3dcba2844d0d8b226f16c284101f,
title = "A Simulator for Concept Detector Output",
abstract = "Concept based video retrieval is a promising search paradigm because it is fully automated and it investigates the fine grained content of a video, which is normally not captured by human annotations. Concepts are captured by so-called concept detectors. However, since these detectors do not yet show a sufficient performance, the evaluation of retrieval systems, which are built on top of the detector output, is difficult. In this report we describe a software package which generates simulated detector output for a specified performance level. Afterwards, this output can be used to execute a search run and ultimately to evaluate the performance of the proposed retrieval method, which is normally done through comparison to a baseline. The probabilistic model of the detectors are two Gaussians, one for the positive and one for the negative class. Thus, the parameters for the simulation are the two means and deviations plus the prior probability of the concept in the dataset.",
keywords = "Detector Simulation, Video Information Retrieval, EWI-16544, IR-68675, METIS-264441",
author = "Robin Aly and Djoerd Hiemstra",
note = "Software download under http://detectsim.sourceforge.net",
year = "2009",
month = "12",
day = "1",
language = "Undefined",
series = "CTIT Technical Report Series",
publisher = "Centre for Telematics and Information Technology (CTIT)",
number = "TR-CTIT-09-40",
address = "Netherlands",

}

Aly, R & Hiemstra, D 2009, A Simulator for Concept Detector Output. CTIT Technical Report Series, no. TR-CTIT-09-40, Centre for Telematics and Information Technology (CTIT), Enschede.

A Simulator for Concept Detector Output. / Aly, Robin; Hiemstra, Djoerd.

Enschede : Centre for Telematics and Information Technology (CTIT), 2009. 12 p. (CTIT Technical Report Series; No. TR-CTIT-09-40).

Research output: Book/ReportReportProfessional

TY - BOOK

T1 - A Simulator for Concept Detector Output

AU - Aly, Robin

AU - Hiemstra, Djoerd

N1 - Software download under http://detectsim.sourceforge.net

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Concept based video retrieval is a promising search paradigm because it is fully automated and it investigates the fine grained content of a video, which is normally not captured by human annotations. Concepts are captured by so-called concept detectors. However, since these detectors do not yet show a sufficient performance, the evaluation of retrieval systems, which are built on top of the detector output, is difficult. In this report we describe a software package which generates simulated detector output for a specified performance level. Afterwards, this output can be used to execute a search run and ultimately to evaluate the performance of the proposed retrieval method, which is normally done through comparison to a baseline. The probabilistic model of the detectors are two Gaussians, one for the positive and one for the negative class. Thus, the parameters for the simulation are the two means and deviations plus the prior probability of the concept in the dataset.

AB - Concept based video retrieval is a promising search paradigm because it is fully automated and it investigates the fine grained content of a video, which is normally not captured by human annotations. Concepts are captured by so-called concept detectors. However, since these detectors do not yet show a sufficient performance, the evaluation of retrieval systems, which are built on top of the detector output, is difficult. In this report we describe a software package which generates simulated detector output for a specified performance level. Afterwards, this output can be used to execute a search run and ultimately to evaluate the performance of the proposed retrieval method, which is normally done through comparison to a baseline. The probabilistic model of the detectors are two Gaussians, one for the positive and one for the negative class. Thus, the parameters for the simulation are the two means and deviations plus the prior probability of the concept in the dataset.

KW - Detector Simulation

KW - Video Information Retrieval

KW - EWI-16544

KW - IR-68675

KW - METIS-264441

M3 - Report

T3 - CTIT Technical Report Series

BT - A Simulator for Concept Detector Output

PB - Centre for Telematics and Information Technology (CTIT)

CY - Enschede

ER -

Aly R, Hiemstra D. A Simulator for Concept Detector Output. Enschede: Centre for Telematics and Information Technology (CTIT), 2009. 12 p. (CTIT Technical Report Series; TR-CTIT-09-40).