Video caption detection algorithm based on multiple instance learning

Haibo Liu; Changjian Zhou; Jing Shen; Pingke Li; Shengping Zhang

doi:10.1109/ICICSE.2010.11

Video caption detection algorithm based on multiple instance learning

Haibo Liu^*
, Changjian Zhou
, Jing Shen
, Pingke Li
, Shengping Zhang

^*Corresponding author for this work

College of Computer Science and Technology, Harbin Engineering University
School of Computer Science and Technology, Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

Over the last few decades, multiple-instance learning (MIL) has been successfully utilized to solve the content-based image/video retrieval (CBIR/CBVR) problem, in which a bag corresponds to a video scene and an instance corresponds to a frame caption. However, existing feature representation schemes are not effective enough to use MIL to detect video caption frames from news video, which hinders the practical applications of CBVR. This paper presents an algorithm that regards the video frames containing caption as a bag. It detects, localizes and extracts video caption frames using multiple-instance learning (MIL) automatically. Experimental results show that the proposed method can detect, localize, and extract video caption frames with more high accuracy.

Original language	English
Title of host publication	Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010
Publisher	IEEE Computer Society
Pages	20-24
Number of pages	5
ISBN (Print)	9780769543390
DOIs	https://doi.org/10.1109/ICICSE.2010.11
State	Published - 2010
Externally published	Yes
Event	5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010 - Harbin, Hei Long Jiang, China Duration: 1 Nov 2010 → 1 Nov 2010

Publication series

Name	Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010

Conference

Conference	5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010
Country/Territory	China
City	Harbin, Hei Long Jiang
Period	1/11/10 → 1/11/10

Keywords

CBVR
Caption detection
MIL
Multi-instance learning
News video retrieval
VideoCaption

Access to Document

10.1109/ICICSE.2010.11

Cite this

Liu, H., Zhou, C., Shen, J., Li, P., & Zhang, S. (2010). Video caption detection algorithm based on multiple instance learning. In Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010 (pp. 20-24). Article 6076534 (Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010). IEEE Computer Society. https://doi.org/10.1109/ICICSE.2010.11

Liu, Haibo ; Zhou, Changjian ; Shen, Jing et al. / Video caption detection algorithm based on multiple instance learning. Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010. IEEE Computer Society, 2010. pp. 20-24 (Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010).

@inproceedings{fd1cea3de852429f82d3d8ed50e55cb2,

title = "Video caption detection algorithm based on multiple instance learning",

abstract = "Over the last few decades, multiple-instance learning (MIL) has been successfully utilized to solve the content-based image/video retrieval (CBIR/CBVR) problem, in which a bag corresponds to a video scene and an instance corresponds to a frame caption. However, existing feature representation schemes are not effective enough to use MIL to detect video caption frames from news video, which hinders the practical applications of CBVR. This paper presents an algorithm that regards the video frames containing caption as a bag. It detects, localizes and extracts video caption frames using multiple-instance learning (MIL) automatically. Experimental results show that the proposed method can detect, localize, and extract video caption frames with more high accuracy.",

keywords = "CBVR, Caption detection, MIL, Multi-instance learning, News video retrieval, VideoCaption",

author = "Haibo Liu and Changjian Zhou and Jing Shen and Pingke Li and Shengping Zhang",

year = "2010",

doi = "10.1109/ICICSE.2010.11",

language = "英语",

isbn = "9780769543390",

series = "Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010",

publisher = "IEEE Computer Society",

pages = "20--24",

booktitle = "Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010",

address = "美国",

note = "5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010 ; Conference date: 01-11-2010 Through 01-11-2010",

}

Liu, H, Zhou, C, Shen, J, Li, P & Zhang, S 2010, Video caption detection algorithm based on multiple instance learning. in Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010., 6076534, Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010, IEEE Computer Society, pp. 20-24, 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010, Harbin, Hei Long Jiang, China, 1/11/10. https://doi.org/10.1109/ICICSE.2010.11

Video caption detection algorithm based on multiple instance learning. / Liu, Haibo; Zhou, Changjian; Shen, Jing et al.
Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010. IEEE Computer Society, 2010. p. 20-24 6076534 (Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Video caption detection algorithm based on multiple instance learning

AU - Liu, Haibo

AU - Zhou, Changjian

AU - Shen, Jing

AU - Li, Pingke

AU - Zhang, Shengping

PY - 2010

Y1 - 2010

N2 - Over the last few decades, multiple-instance learning (MIL) has been successfully utilized to solve the content-based image/video retrieval (CBIR/CBVR) problem, in which a bag corresponds to a video scene and an instance corresponds to a frame caption. However, existing feature representation schemes are not effective enough to use MIL to detect video caption frames from news video, which hinders the practical applications of CBVR. This paper presents an algorithm that regards the video frames containing caption as a bag. It detects, localizes and extracts video caption frames using multiple-instance learning (MIL) automatically. Experimental results show that the proposed method can detect, localize, and extract video caption frames with more high accuracy.

AB - Over the last few decades, multiple-instance learning (MIL) has been successfully utilized to solve the content-based image/video retrieval (CBIR/CBVR) problem, in which a bag corresponds to a video scene and an instance corresponds to a frame caption. However, existing feature representation schemes are not effective enough to use MIL to detect video caption frames from news video, which hinders the practical applications of CBVR. This paper presents an algorithm that regards the video frames containing caption as a bag. It detects, localizes and extracts video caption frames using multiple-instance learning (MIL) automatically. Experimental results show that the proposed method can detect, localize, and extract video caption frames with more high accuracy.

KW - CBVR

KW - Caption detection

KW - MIL

KW - Multi-instance learning

KW - News video retrieval

KW - VideoCaption

UR - https://www.scopus.com/pages/publications/82955171404

U2 - 10.1109/ICICSE.2010.11

DO - 10.1109/ICICSE.2010.11

M3 - 会议稿件

AN - SCOPUS:82955171404

SN - 9780769543390

T3 - Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010

SP - 20

EP - 24

BT - Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010

PB - IEEE Computer Society

T2 - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010

Y2 - 1 November 2010 through 1 November 2010

ER -

Liu H, Zhou C, Shen J, Li P, Zhang S. Video caption detection algorithm based on multiple instance learning. In Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010. IEEE Computer Society. 2010. p. 20-24. 6076534. (Proceedings - 5th International Conference on Internet Computing for Science and Engineering, ICICSE 2010). doi: 10.1109/ICICSE.2010.11

Video caption detection algorithm based on multiple instance learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this