Effective moving object detection and retrieval via integrating spatial-temporal multimedia information

Dianting Liu, Mei-Ling Shyu

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

7 Citations (Scopus)

Abstract

In the area of multimedia semantic analysis and video retrieval, automatic object detection techniques play an important role. Without the analysis of the object-level features, it is hard to achieve high performance on semantic retrieval. As a branch of object detection study, moving object detection also becomes a hot research field and gets a great amount of progress recently. This paper proposes a moving object detection and retrieval model that integrates the spatial and temporal information in video sequences and uses the proposed integral density method (adopted from the idea of integral images) to quickly identify the motion regions in an unsupervised way. First, key information locations on video frames are achieved as maxima and minima of the result of Difference of Gaussian (DoG) function. On the other hand, a motion map of adjacent frames is obtained from the diversity of the outcomes from Simultaneous Partition and Class Parameter Estimation (SPCPE) framework. The motion map filters key information locations into key motion locations (KMLs) where the existence of moving objects is implied. Besides showing the motion zones, the motion map also indicates the motion direction which guides the proposed "integral density" approach to quickly and accurately locate the motion regions. The detection results are not only illustrated visually, but also verified by the promising experimental results which show the concept retrieval performance can be improved by integrating the global and local visual information.

Original languageEnglish
Title of host publicationProceedings - 2012 IEEE International Symposium on Multimedia, ISM 2012
Pages364-371
Number of pages8
DOIs
StatePublished - Dec 1 2012
Event14th IEEE International Symposium on Multimedia, ISM 2012 - Irvine, CA, United States
Duration: Dec 10 2012Dec 12 2012

Other

Other14th IEEE International Symposium on Multimedia, ISM 2012
CountryUnited States
CityIrvine, CA
Period12/10/1212/12/12

Fingerprint

Semantics
Parameter estimation
Object detection

Keywords

  • Integral density
  • Integral image
  • Key motion location
  • Moving object
  • Spatial-temporal
  • SPCPE

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Liu, D., & Shyu, M-L. (2012). Effective moving object detection and retrieval via integrating spatial-temporal multimedia information. In Proceedings - 2012 IEEE International Symposium on Multimedia, ISM 2012 (pp. 364-371). [6424688] https://doi.org/10.1109/ISM.2012.74

Effective moving object detection and retrieval via integrating spatial-temporal multimedia information. / Liu, Dianting; Shyu, Mei-Ling.

Proceedings - 2012 IEEE International Symposium on Multimedia, ISM 2012. 2012. p. 364-371 6424688.

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

Liu, D & Shyu, M-L 2012, Effective moving object detection and retrieval via integrating spatial-temporal multimedia information. in Proceedings - 2012 IEEE International Symposium on Multimedia, ISM 2012., 6424688, pp. 364-371, 14th IEEE International Symposium on Multimedia, ISM 2012, Irvine, CA, United States, 12/10/12. https://doi.org/10.1109/ISM.2012.74
Liu D, Shyu M-L. Effective moving object detection and retrieval via integrating spatial-temporal multimedia information. In Proceedings - 2012 IEEE International Symposium on Multimedia, ISM 2012. 2012. p. 364-371. 6424688 https://doi.org/10.1109/ISM.2012.74
Liu, Dianting ; Shyu, Mei-Ling. / Effective moving object detection and retrieval via integrating spatial-temporal multimedia information. Proceedings - 2012 IEEE International Symposium on Multimedia, ISM 2012. 2012. pp. 364-371
@inproceedings{92d404bb43ba48e6bd74884ee768e633,
title = "Effective moving object detection and retrieval via integrating spatial-temporal multimedia information",
abstract = "In the area of multimedia semantic analysis and video retrieval, automatic object detection techniques play an important role. Without the analysis of the object-level features, it is hard to achieve high performance on semantic retrieval. As a branch of object detection study, moving object detection also becomes a hot research field and gets a great amount of progress recently. This paper proposes a moving object detection and retrieval model that integrates the spatial and temporal information in video sequences and uses the proposed integral density method (adopted from the idea of integral images) to quickly identify the motion regions in an unsupervised way. First, key information locations on video frames are achieved as maxima and minima of the result of Difference of Gaussian (DoG) function. On the other hand, a motion map of adjacent frames is obtained from the diversity of the outcomes from Simultaneous Partition and Class Parameter Estimation (SPCPE) framework. The motion map filters key information locations into key motion locations (KMLs) where the existence of moving objects is implied. Besides showing the motion zones, the motion map also indicates the motion direction which guides the proposed {"}integral density{"} approach to quickly and accurately locate the motion regions. The detection results are not only illustrated visually, but also verified by the promising experimental results which show the concept retrieval performance can be improved by integrating the global and local visual information.",
keywords = "Integral density, Integral image, Key motion location, Moving object, Spatial-temporal, SPCPE",
author = "Dianting Liu and Mei-Ling Shyu",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/ISM.2012.74",
language = "English",
isbn = "9780769548753",
pages = "364--371",
booktitle = "Proceedings - 2012 IEEE International Symposium on Multimedia, ISM 2012",

}

TY - GEN

T1 - Effective moving object detection and retrieval via integrating spatial-temporal multimedia information

AU - Liu, Dianting

AU - Shyu, Mei-Ling

PY - 2012/12/1

Y1 - 2012/12/1

N2 - In the area of multimedia semantic analysis and video retrieval, automatic object detection techniques play an important role. Without the analysis of the object-level features, it is hard to achieve high performance on semantic retrieval. As a branch of object detection study, moving object detection also becomes a hot research field and gets a great amount of progress recently. This paper proposes a moving object detection and retrieval model that integrates the spatial and temporal information in video sequences and uses the proposed integral density method (adopted from the idea of integral images) to quickly identify the motion regions in an unsupervised way. First, key information locations on video frames are achieved as maxima and minima of the result of Difference of Gaussian (DoG) function. On the other hand, a motion map of adjacent frames is obtained from the diversity of the outcomes from Simultaneous Partition and Class Parameter Estimation (SPCPE) framework. The motion map filters key information locations into key motion locations (KMLs) where the existence of moving objects is implied. Besides showing the motion zones, the motion map also indicates the motion direction which guides the proposed "integral density" approach to quickly and accurately locate the motion regions. The detection results are not only illustrated visually, but also verified by the promising experimental results which show the concept retrieval performance can be improved by integrating the global and local visual information.

AB - In the area of multimedia semantic analysis and video retrieval, automatic object detection techniques play an important role. Without the analysis of the object-level features, it is hard to achieve high performance on semantic retrieval. As a branch of object detection study, moving object detection also becomes a hot research field and gets a great amount of progress recently. This paper proposes a moving object detection and retrieval model that integrates the spatial and temporal information in video sequences and uses the proposed integral density method (adopted from the idea of integral images) to quickly identify the motion regions in an unsupervised way. First, key information locations on video frames are achieved as maxima and minima of the result of Difference of Gaussian (DoG) function. On the other hand, a motion map of adjacent frames is obtained from the diversity of the outcomes from Simultaneous Partition and Class Parameter Estimation (SPCPE) framework. The motion map filters key information locations into key motion locations (KMLs) where the existence of moving objects is implied. Besides showing the motion zones, the motion map also indicates the motion direction which guides the proposed "integral density" approach to quickly and accurately locate the motion regions. The detection results are not only illustrated visually, but also verified by the promising experimental results which show the concept retrieval performance can be improved by integrating the global and local visual information.

KW - Integral density

KW - Integral image

KW - Key motion location

KW - Moving object

KW - Spatial-temporal

KW - SPCPE

UR - http://www.scopus.com/inward/record.url?scp=84874266857&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874266857&partnerID=8YFLogxK

U2 - 10.1109/ISM.2012.74

DO - 10.1109/ISM.2012.74

M3 - Conference contribution

SN - 9780769548753

SP - 364

EP - 371

BT - Proceedings - 2012 IEEE International Symposium on Multimedia, ISM 2012

ER -