A unified shot boundary detection method based on linear prediction with bayesian cost functions

Cheng Cai; Kin Man Lam; Zheng Tan

doi:10.1109/IWVDVT.2005.1504561

A unified shot boundary detection method based on linear prediction with bayesian cost functions

Cheng Cai
, Kin Man Lam
, Zheng Tan

The Hong Kong Polytechnic University

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

7 Citations (Scopus)

Abstract

The detection of edits in a video sequence is the first step for video analysis, which segments a video into its basic components. In this paper, we propose a novel and efficient approach for shot boundary detection, which can detect cuts and dissolves reliably using a uniform framework. Our approach is based on the temporal linear prediction of the frames. With frames in a video shot, a frame can be predicted from its previous frames. If the prediction error is high, a cut should happen. For dissolves, the gradual transitions can be modeled by temporal linear prediction with constant prediction coefficients. Experimental results show that our algorithm can achieve high precision even if a video contains object motion and camera motion, and is able to detect and classify cuts and dissolves in real time.

Original language	English
Title of host publication	Proceedings of the 2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005
Publisher	IEEE
Pages	163-166
Number of pages	4
ISBN (Print)	0780390067
DOIs	https://doi.org/10.1109/IWVDVT.2005.1504561
Publication status	Published - 25 Aug 2005
Event	2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005 - Suzhou, China Duration: 28 May 2005 → 30 May 2005

Conference

Conference	2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005
Country/Territory	China
City	Suzhou
Period	28/05/05 → 30/05/05

Keywords

Bayes methods
Image segmentation
Video signal processing

ASJC Scopus subject areas

General Engineering

More information

10.1109/IWVDVT.2005.1504561

Cite this

@inproceedings{5eeec5aff6ef4c6cbf0c2a68fbe97da9,

title = "A unified shot boundary detection method based on linear prediction with bayesian cost functions",

abstract = "The detection of edits in a video sequence is the first step for video analysis, which segments a video into its basic components. In this paper, we propose a novel and efficient approach for shot boundary detection, which can detect cuts and dissolves reliably using a uniform framework. Our approach is based on the temporal linear prediction of the frames. With frames in a video shot, a frame can be predicted from its previous frames. If the prediction error is high, a cut should happen. For dissolves, the gradual transitions can be modeled by temporal linear prediction with constant prediction coefficients. Experimental results show that our algorithm can achieve high precision even if a video contains object motion and camera motion, and is able to detect and classify cuts and dissolves in real time.",

keywords = "Bayes methods, Image segmentation, Video signal processing",

author = "Cheng Cai and Lam, \{Kin Man\} and Zheng Tan",

year = "2005",

month = aug,

day = "25",

doi = "10.1109/IWVDVT.2005.1504561",

language = "English",

isbn = "0780390067",

pages = "163--166",

booktitle = "Proceedings of the 2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005",

publisher = "IEEE",

address = "United States",

note = "2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005 ; Conference date: 28-05-2005 Through 30-05-2005",

}

Cai, C, Lam, KM & Tan, Z 2005, A unified shot boundary detection method based on linear prediction with bayesian cost functions. in Proceedings of the 2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005. IEEE, pp. 163-166, 2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005, Suzhou, China, 28/05/05. https://doi.org/10.1109/IWVDVT.2005.1504561

A unified shot boundary detection method based on linear prediction with bayesian cost functions. / Cai, Cheng; Lam, Kin Man; Tan, Zheng.
Proceedings of the 2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005. IEEE, 2005. p. 163-166.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - A unified shot boundary detection method based on linear prediction with bayesian cost functions

AU - Cai, Cheng

AU - Lam, Kin Man

AU - Tan, Zheng

PY - 2005/8/25

Y1 - 2005/8/25

N2 - The detection of edits in a video sequence is the first step for video analysis, which segments a video into its basic components. In this paper, we propose a novel and efficient approach for shot boundary detection, which can detect cuts and dissolves reliably using a uniform framework. Our approach is based on the temporal linear prediction of the frames. With frames in a video shot, a frame can be predicted from its previous frames. If the prediction error is high, a cut should happen. For dissolves, the gradual transitions can be modeled by temporal linear prediction with constant prediction coefficients. Experimental results show that our algorithm can achieve high precision even if a video contains object motion and camera motion, and is able to detect and classify cuts and dissolves in real time.

AB - The detection of edits in a video sequence is the first step for video analysis, which segments a video into its basic components. In this paper, we propose a novel and efficient approach for shot boundary detection, which can detect cuts and dissolves reliably using a uniform framework. Our approach is based on the temporal linear prediction of the frames. With frames in a video shot, a frame can be predicted from its previous frames. If the prediction error is high, a cut should happen. For dissolves, the gradual transitions can be modeled by temporal linear prediction with constant prediction coefficients. Experimental results show that our algorithm can achieve high precision even if a video contains object motion and camera motion, and is able to detect and classify cuts and dissolves in real time.

KW - Bayes methods

KW - Image segmentation

KW - Video signal processing

UR - https://www.scopus.com/pages/publications/23844557593

U2 - 10.1109/IWVDVT.2005.1504561

DO - 10.1109/IWVDVT.2005.1504561

M3 - Conference article published in proceeding or book

SN - 0780390067

SP - 163

EP - 166

BT - Proceedings of the 2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005

PB - IEEE

T2 - 2005 IEEE International Workshop on VLSI Design and Video Technology, IWVDVT 2005

Y2 - 28 May 2005 through 30 May 2005

ER -

A unified shot boundary detection method based on linear prediction with bayesian cost functions

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this