Automatic semantic annotation of images using spatial hidden Markov model

Feiyang Yu; Horace H. S. Ip

doi:10.1109/ICME.2006.262459

Automatic semantic annotation of images using spatial hidden Markov model

Feiyang Yu
, Horace H. S. Ip

Department of Computer Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

18 Citations (Scopus)

Abstract

This paper presents a new spatial-HMM(SHMM)for automatically classifying and annotating natural images. Our model is a 2D generalization of the traditional HMM in the sense that both vertical and horizontal transitions between hidden states are taken into consideration. The three basic problems with HMM-liked model are also solved in our model. Given a sequence of visual features, our model automatically derives annotations from keywords associated with the most appropriate concept class, and with no need of a pre-defined length threshold. Our experiments showed that our model outperformed the previous 2D MHMM in recognition accuracy and also achieved a high annotation accuracy. © 2006 IEEE.

Original language	English
Title of host publication	2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings
Pages	305-308
Volume	2006
DOIs	https://doi.org/10.1109/ICME.2006.262459
Publication status	Published - 2006
Event	2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Toronto, Canada Duration: 9 Jul 2006 → 12 Jul 2006 https://ieeexplore.ieee.org/document/1576931

Publication series

Name
Volume	2006

Conference

Conference	2006 IEEE International Conference on Multimedia and Expo, ICME 2006
Place	Canada
City	Toronto
Period	9/07/06 → 12/07/06
Internet address	https://ieeexplore.ieee.org/document/1576931

Access Output

10.1109/ICME.2006.262459

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{3a63edf65b8a473b8a032b6fb271f40b,

title = "Automatic semantic annotation of images using spatial hidden Markov model",

abstract = "This paper presents a new spatial-HMM(SHMM)for automatically classifying and annotating natural images. Our model is a 2D generalization of the traditional HMM in the sense that both vertical and horizontal transitions between hidden states are taken into consideration. The three basic problems with HMM-liked model are also solved in our model. Given a sequence of visual features, our model automatically derives annotations from keywords associated with the most appropriate concept class, and with no need of a pre-defined length threshold. Our experiments showed that our model outperformed the previous 2D MHMM in recognition accuracy and also achieved a high annotation accuracy. {\textcopyright} 2006 IEEE.",

author = "Feiyang Yu and Ip, \{Horace H. S.\}",

year = "2006",

doi = "10.1109/ICME.2006.262459",

language = "English",

isbn = "1424403677",

volume = "2006",

pages = "305--308",

booktitle = "2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings",

note = "2006 IEEE International Conference on Multimedia and Expo, ICME 2006 ; Conference date: 09-07-2006 Through 12-07-2006",

url = "https://ieeexplore.ieee.org/document/1576931",

}

Automatic semantic annotation of images using spatial hidden Markov model. / Yu, Feiyang; Ip, Horace H. S.
2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings. Vol. 2006 2006. p. 305-308 4036597.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Automatic semantic annotation of images using spatial hidden Markov model

AU - Yu, Feiyang

AU - Ip, Horace H. S.

PY - 2006

Y1 - 2006

N2 - This paper presents a new spatial-HMM(SHMM)for automatically classifying and annotating natural images. Our model is a 2D generalization of the traditional HMM in the sense that both vertical and horizontal transitions between hidden states are taken into consideration. The three basic problems with HMM-liked model are also solved in our model. Given a sequence of visual features, our model automatically derives annotations from keywords associated with the most appropriate concept class, and with no need of a pre-defined length threshold. Our experiments showed that our model outperformed the previous 2D MHMM in recognition accuracy and also achieved a high annotation accuracy. © 2006 IEEE.

AB - This paper presents a new spatial-HMM(SHMM)for automatically classifying and annotating natural images. Our model is a 2D generalization of the traditional HMM in the sense that both vertical and horizontal transitions between hidden states are taken into consideration. The three basic problems with HMM-liked model are also solved in our model. Given a sequence of visual features, our model automatically derives annotations from keywords associated with the most appropriate concept class, and with no need of a pre-defined length threshold. Our experiments showed that our model outperformed the previous 2D MHMM in recognition accuracy and also achieved a high annotation accuracy. © 2006 IEEE.

UR - https://www.scopus.com/pages/publications/34247636308

UR - https://www.scopus.com/pages/publications/34247636308#tab=impact

U2 - 10.1109/ICME.2006.262459

DO - 10.1109/ICME.2006.262459

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 1424403677

SN - 9781424403677

VL - 2006

SP - 305

EP - 308

BT - 2006 IEEE International Conference on Multimedia and Expo, ICME 2006 - Proceedings

T2 - 2006 IEEE International Conference on Multimedia and Expo, ICME 2006

Y2 - 9 July 2006 through 12 July 2006

ER -

Automatic semantic annotation of images using spatial hidden Markov model

Abstract

Publication series

Conference

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this