Statistical analysis on manifolds and its applications to video analysis

Pavan Turaga; Ashok Veeraraghavan; Anuj Srivastava; Rama Chellappa

doi:10.1007/978-3-642-12900-1_5

Statistical analysis on manifolds and its applications to video analysis

Pavan Turaga, Ashok Veeraraghavan, Anuj Srivastava, Rama Chellappa

Research output: Chapter in Book/Report/Conference proceeding › Chapter

11 Scopus citations

Abstract

The analysis and interpretation of video data is an important component of modern vision applications such as biometrics, surveillance, motionsynthesis and web-based user interfaces. A common requirement among these very different applications is the ability to learn statistical models of appearance and motion from a collection of videos, and then use them for recognizing actions or persons in a new video. These applications in video analysis require statistical inference methods to be devised on non-Euclidean spaces or more formally on manifolds. This chapter outlines a broad survey of applications in video analysis that involve manifolds. We develop the required mathematical tools needed to perform statistical inference on manifolds and show their effectiveness in real video-understanding applications.

Original language	English (US)
Title of host publication	Video Search and Mining
Editors	Dan Schonfeld, Caifeng Shan, Dacheng Tao, Liang Wang
Pages	115-144
Number of pages	30
DOIs	https://doi.org/10.1007/978-3-642-12900-1_5
State	Published - 2010
Externally published	Yes

Publication series

Name	Studies in Computational Intelligence
Volume	287
ISSN (Print)	1860-949X

ASJC Scopus subject areas

Artificial Intelligence

Access to Document

10.1007/978-3-642-12900-1_5

Cite this

@inbook{de76ccdd13204677a390b3e3cc3847bc,

title = "Statistical analysis on manifolds and its applications to video analysis",

abstract = "The analysis and interpretation of video data is an important component of modern vision applications such as biometrics, surveillance, motionsynthesis and web-based user interfaces. A common requirement among these very different applications is the ability to learn statistical models of appearance and motion from a collection of videos, and then use them for recognizing actions or persons in a new video. These applications in video analysis require statistical inference methods to be devised on non-Euclidean spaces or more formally on manifolds. This chapter outlines a broad survey of applications in video analysis that involve manifolds. We develop the required mathematical tools needed to perform statistical inference on manifolds and show their effectiveness in real video-understanding applications.",

author = "Pavan Turaga and Ashok Veeraraghavan and Anuj Srivastava and Rama Chellappa",

year = "2010",

doi = "10.1007/978-3-642-12900-1_5",

language = "English (US)",

isbn = "9783642128998",

series = "Studies in Computational Intelligence",

pages = "115--144",

editor = "Dan Schonfeld and Caifeng Shan and Dacheng Tao and Liang Wang",

booktitle = "Video Search and Mining",

}

TY - CHAP

T1 - Statistical analysis on manifolds and its applications to video analysis

AU - Turaga, Pavan

AU - Veeraraghavan, Ashok

AU - Srivastava, Anuj

AU - Chellappa, Rama

PY - 2010

Y1 - 2010

N2 - The analysis and interpretation of video data is an important component of modern vision applications such as biometrics, surveillance, motionsynthesis and web-based user interfaces. A common requirement among these very different applications is the ability to learn statistical models of appearance and motion from a collection of videos, and then use them for recognizing actions or persons in a new video. These applications in video analysis require statistical inference methods to be devised on non-Euclidean spaces or more formally on manifolds. This chapter outlines a broad survey of applications in video analysis that involve manifolds. We develop the required mathematical tools needed to perform statistical inference on manifolds and show their effectiveness in real video-understanding applications.

AB - The analysis and interpretation of video data is an important component of modern vision applications such as biometrics, surveillance, motionsynthesis and web-based user interfaces. A common requirement among these very different applications is the ability to learn statistical models of appearance and motion from a collection of videos, and then use them for recognizing actions or persons in a new video. These applications in video analysis require statistical inference methods to be devised on non-Euclidean spaces or more formally on manifolds. This chapter outlines a broad survey of applications in video analysis that involve manifolds. We develop the required mathematical tools needed to perform statistical inference on manifolds and show their effectiveness in real video-understanding applications.

UR - http://www.scopus.com/inward/record.url?scp=77951712933&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951712933&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-12900-1_5

DO - 10.1007/978-3-642-12900-1_5

M3 - Chapter

AN - SCOPUS:77951712933

SN - 9783642128998

T3 - Studies in Computational Intelligence

SP - 115

EP - 144

BT - Video Search and Mining

A2 - Schonfeld, Dan

A2 - Shan, Caifeng

A2 - Tao, Dacheng

A2 - Wang, Liang

ER -

Statistical analysis on manifolds and its applications to video analysis

Abstract

Publication series

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this