Motion Primitives and Probabilistic Edit Distance for Action Recognition

Preben Fihl; Michael Boelstoft Holte; Thomas B. Moeslund

doi:10.1007/978-3-540-92865-2_3

Motion Primitives and Probabilistic Edit Distance for Action Recognition

Preben Fihl, Michael Boelstoft Holte, Thomas B. Moeslund

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › peer review

Abstract

Udgivelsesdato: January 14

Originalsprog	Engelsk
Bogserie	Lecture Notes in Computer Science
Vol/bind	5085
Udgave nummer	1
Sider (fra-til)	24-35
Antal sider	12
ISSN	0302-9743
DOI	https://doi.org/10.1007/978-3-540-92865-2_3
Status	Udgivet - 2009

Adgang til dokumentet

10.1007/978-3-540-92865-2_3

AUB Link

Søg efter materialet i Aalborg Universitetsbiblioteks søgemaskine

HERMES - Human-Expressive Representations of Motion and Their Evaluation in Sequences
Fihl, P., Holte, M. B., Reng, L., Moeslund, T. B. & Granum, E.
<ingen navn>
01/04/2006 → 31/08/2009
Projekter: Projekt › Forskning
MoPrim - In search for the motion primitives for a communicative human body language / MoPrim - Bevægelsesprimitiver for det menneskelige kropssprog
Moeslund, T. B., Reng, L. & Fihl, P.
01/04/2004 → 01/05/2007
Projekter: Projekt › Forskning

Citationsformater

@article{0f5bc570eea511ddb0a4000ea68e967b,

title = "Motion Primitives and Probabilistic Edit Distance for Action Recognition",

abstract = "The number of potential applications has made automatic recognition of human actions a very active research area. Different approaches have been followed based on trajectories through some state space. In this paper we also model an action as a trajectory through a state space, but we represent the actions as a sequence of temporal isolated instances, denoted primitives. These primitives are each defined by four features extracted from motion images. The primitives are recognized in each frame based on a trained classifier resulting in a sequence of primitives. From this sequence we recognize different temporal actions using a probabilistic Edit Distance method. The method is tested on different actions with and without noise and the results show recognition rates of 88.7% and 85.5%, respectively.",

keywords = "Computer vision, Human action recognition, Motion primitives",

author = "Preben Fihl and Holte, {Michael Boelstoft} and Moeslund, {Thomas B.}",

year = "2009",

doi = "10.1007/978-3-540-92865-2_3",

language = "English",

volume = "5085",

pages = "24--35",

journal = "Lecture Notes in Computer Science",

issn = "0302-9743",

publisher = "Physica-Verlag",

number = "1",

}

TY - JOUR

T1 - Motion Primitives and Probabilistic Edit Distance for Action Recognition

AU - Fihl, Preben

AU - Holte, Michael Boelstoft

AU - Moeslund, Thomas B.

PY - 2009

Y1 - 2009

N2 - The number of potential applications has made automatic recognition of human actions a very active research area. Different approaches have been followed based on trajectories through some state space. In this paper we also model an action as a trajectory through a state space, but we represent the actions as a sequence of temporal isolated instances, denoted primitives. These primitives are each defined by four features extracted from motion images. The primitives are recognized in each frame based on a trained classifier resulting in a sequence of primitives. From this sequence we recognize different temporal actions using a probabilistic Edit Distance method. The method is tested on different actions with and without noise and the results show recognition rates of 88.7% and 85.5%, respectively.

AB - The number of potential applications has made automatic recognition of human actions a very active research area. Different approaches have been followed based on trajectories through some state space. In this paper we also model an action as a trajectory through a state space, but we represent the actions as a sequence of temporal isolated instances, denoted primitives. These primitives are each defined by four features extracted from motion images. The primitives are recognized in each frame based on a trained classifier resulting in a sequence of primitives. From this sequence we recognize different temporal actions using a probabilistic Edit Distance method. The method is tested on different actions with and without noise and the results show recognition rates of 88.7% and 85.5%, respectively.

KW - Computer vision

KW - Human action recognition

KW - Motion primitives

U2 - 10.1007/978-3-540-92865-2_3

DO - 10.1007/978-3-540-92865-2_3

M3 - Journal article

SN - 0302-9743

VL - 5085

SP - 24

EP - 35

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

IS - 1

ER -

Motion Primitives and Probabilistic Edit Distance for Action Recognition

Abstract

Adgang til dokumentet

AUB Link

Fingeraftryk

Projekter

HERMES - Human-Expressive Representations of Motion and Their Evaluation in Sequences

MoPrim - In search for the motion primitives for a communicative human body language / MoPrim - Bevægelsesprimitiver for det menneskelige kropssprog

Citationsformater