Detection of unique people in news programs using multimodal shot clustering

Get BibTex-formatted data

Download

PDF

Author

CM Taskiran, A Albiol, L Torres, EJ Delp

Entry type

article

Abstract

In this paper, we describe an approach that uses a combination of visual and audio features to cluster shots belonging to the same person in video programs. We use color histograms extracted from keyframes and faces, as well as cepstral coefficients derived from audio to calculate pairwise shot distances. These distances are then normalized and combined to a single confidence value which reflects our certainty that two shots contain the same person. We then use an agglomerative clustering algorithm to cluster shots based on these confidence values. We report the results of our system on a data set of approximately 8 hours of programming.

Download

PDF

Date

2004 – 10

URL

http://ieeexplore.ieee.org/iel ... ber=&arnumber=1418850

Journal

Image Processing, 2004. ICIP '04. 2004 International Conference on

Key alpha

Delp

Pages

697-700

Volume

Publication Date

2004-10-01

BibTex-formatted data

To refer to this entry, you may select and copy the text below and paste it into your BibTex document. Note that the text may not contain all macros that BibTex supports.

@Article{ Delp,
	title = "Detection of unique people in news programs using multimodal shot clustering",
	author = "CM Taskiran, A Albiol, L Torres, EJ Delp",
	year = "2004",
	month = "10",
	journal = "Image Processing, 2004. ICIP '04. 2004 International Conference on",
	pages = "697-700",
	volume = "1",
	abstract = "In this paper, we describe an approach that uses a combination of visual and audio features to cluster shots belonging to the same person in video programs. We use color histograms extracted from keyframes and faces, as well as cepstral coefficients derived from audio to calculate pairwise shot distances. These distances are then normalized and combined to a single confidence value which reflects our certainty that two shots contain the same person. We then use an agglomerative clustering algorithm to cluster shots based on these confidence values. We report the results of our system on a data set of approximately 8 hours of programming.",
	url = "http://ieeexplore.ieee.org/iel5/9716/30672/01418850.pdf?tp=&isnumber=&arnumber=1418850",
}