Weighted Block Matching-Based Anchor Shot Detection with Dynamic Background

Authors:
Fuguang Zheng;Shijin Li;Hao Li;Jun Feng
Affiliations:
School of Computer & Information Engineering, Hohai University, Nanjing, China;School of Computer & Information Engineering, Hohai University, Nanjing, China;School of Computer & Information Engineering, Hohai University, Nanjing, China;School of Computer & Information Engineering, Hohai University, Nanjing, China
Venue:
ICIAR '09 Proceedings of the 6th International Conference on Image Analysis and Recognition
Year:
2009

Citing 6
Cited 1

Automatic partitioning of full-motion video

Multimedia Systems
Multi-level Anchorperson Detection Using Multimodal Association

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 3 - Volume 03
Anchorperson extraction for picture in picture news video

Pattern Recognition Letters
An Unsupervised Algorithm for Anchor Shot Detection

ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 02
A Multi-Stage Approach for News Video Segmentation Based on Automatic Anchorperson Number Detection

UBICOMM '07 Proceedings of the International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies
A temporal and visual analysis-based approach to commercial detection in news video

VISUAL'07 Proceedings of the 9th international conference on Advances in visual information systems

Anchor shot detection with diverse style backgrounds based on spatial-temporal slice analysis

MMM'10 Proceedings of the 16th international conference on Advances in Multimedia Modeling

Quantified Score

Hi-index	0.00

Visualization

Abstract

Anchor shot detection is fundamental to the structure analysis of news program, such as story segmentation and commercial block location. Previous work mainly tackled the problem of picture in frame on half screen, while there are also some news programs which have a global dynamic studio background. In this paper, a new block weighing based anchor shot detection algorithm is proposed. Firstly, the key frame of each shot is extracted and divided into 64 blocks equally. Then a weighted template is built automatically after clustering analysis of different types of shots, which indicates the degrees of variation on the different blocks in the anchor shot background. And lastly, all anchor shots are detected by a weighted template matching scheme. The detection results have shown that our method is effective, especially suitable for the news program which has a global dynamic background.