Pose Sentences: A new representation for action recognition using sequence of pose words

Kardelen Hatun; Pinar Duygulu

doi:10.1109/ICPR.2008.4761702

Pose Sentences: A new representation for action recognition using sequence of pose words

Kardelen Hatun, Pinar Duygulu

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

18 Citations (Scopus)

177 Downloads (Pure)

Abstract

We propose a method for recognizing human actions in videos. Inspired from the recent bag-of-words approaches, we represent actions as documents consisting of words, where a word refers to the pose in a frame. Histogram of oriented gradients (HOG) features are used to describe poses, which are then vector quantized to obtain pose-words. As an alternative to bag-of-words approaches, that only represent actions as a collection of words by discarding the temporal characteristics of actions, we represent videos as ordered sequence of pose-words, that is as pose sentences. Then, string matching techniques are exploited to find the similarity of two action sequences. In the experiments, performed on data set of Blank et al., 92% performance is obtained.

Original language	Undefined
Title of host publication	International Conference on Pattern Recognition
Place of Publication	Piscataway
Publisher	IEEE
Pages	-
Number of pages	4
ISBN (Print)	978-1-4244-2175-6
DOIs	https://doi.org/10.1109/ICPR.2008.4761702
Publication status	Published - Dec 2008
Event	19th International Conference on Pattern Recognition 2008 - Tampa Convention Center, Tampa, United States Duration: 8 Dec 2008 → 11 Dec 2008 Conference number: 19 https://www.ieee.org/conferences_events/conferences/conferencedetails/index.html?Conf_ID=13355

Publication series

Name
Publisher	IEEE Computer Society Press

Conference

Conference	19th International Conference on Pattern Recognition 2008
Abbreviated title	ICPR 2008
Country/Territory	United States
City	Tampa
Period	8/12/08 → 11/12/08
Other	8 - 11 Dec 2008
Internet address	https://www.ieee.org/conferences_events/conferences/conferencedetails/index.html?Conf_ID=13355

Keywords

EWI-14780
CR-I.4
string matching
METIS-268949
IR-70808
Action recognition
bag-of-words

Access to Document

10.1109/ICPR.2008.4761702

Cite this

@inproceedings{1ce92ce1621244e3b3bd8b35b20e728c,

title = "Pose Sentences: A new representation for action recognition using sequence of pose words",

abstract = "We propose a method for recognizing human actions in videos. Inspired from the recent bag-of-words approaches, we represent actions as documents consisting of words, where a word refers to the pose in a frame. Histogram of oriented gradients (HOG) features are used to describe poses, which are then vector quantized to obtain pose-words. As an alternative to bag-of-words approaches, that only represent actions as a collection of words by discarding the temporal characteristics of actions, we represent videos as ordered sequence of pose-words, that is as pose sentences. Then, string matching techniques are exploited to find the similarity of two action sequences. In the experiments, performed on data set of Blank et al., 92% performance is obtained.",

keywords = "EWI-14780, CR-I.4, string matching, METIS-268949, IR-70808, Action recognition, bag-of-words",

author = "Kardelen Hatun and Pinar Duygulu",

note = "10.1109/ICPR.2008.4761702 ; 19th International Conference on Pattern Recognition 2008 ; Conference date: 08-12-2008 Through 11-12-2008",

year = "2008",

month = dec,

doi = "10.1109/ICPR.2008.4761702",

language = "Undefined",

isbn = "978-1-4244-2175-6",

publisher = "IEEE",

pages = "--",

booktitle = "International Conference on Pattern Recognition",

address = "United States",

url = "https://www.ieee.org/conferences_events/conferences/conferencedetails/index.html?Conf_ID=13355",

}

TY - GEN

T1 - Pose Sentences: A new representation for action recognition using sequence of pose words

AU - Hatun, Kardelen

AU - Duygulu, Pinar

N1 - Conference code: 19

PY - 2008/12

Y1 - 2008/12

N2 - We propose a method for recognizing human actions in videos. Inspired from the recent bag-of-words approaches, we represent actions as documents consisting of words, where a word refers to the pose in a frame. Histogram of oriented gradients (HOG) features are used to describe poses, which are then vector quantized to obtain pose-words. As an alternative to bag-of-words approaches, that only represent actions as a collection of words by discarding the temporal characteristics of actions, we represent videos as ordered sequence of pose-words, that is as pose sentences. Then, string matching techniques are exploited to find the similarity of two action sequences. In the experiments, performed on data set of Blank et al., 92% performance is obtained.

AB - We propose a method for recognizing human actions in videos. Inspired from the recent bag-of-words approaches, we represent actions as documents consisting of words, where a word refers to the pose in a frame. Histogram of oriented gradients (HOG) features are used to describe poses, which are then vector quantized to obtain pose-words. As an alternative to bag-of-words approaches, that only represent actions as a collection of words by discarding the temporal characteristics of actions, we represent videos as ordered sequence of pose-words, that is as pose sentences. Then, string matching techniques are exploited to find the similarity of two action sequences. In the experiments, performed on data set of Blank et al., 92% performance is obtained.

KW - EWI-14780

KW - CR-I.4

KW - string matching

KW - METIS-268949

KW - IR-70808

KW - Action recognition

KW - bag-of-words

U2 - 10.1109/ICPR.2008.4761702

DO - 10.1109/ICPR.2008.4761702

M3 - Conference contribution

SN - 978-1-4244-2175-6

SP - -

BT - International Conference on Pattern Recognition

PB - IEEE

CY - Piscataway

T2 - 19th International Conference on Pattern Recognition 2008

Y2 - 8 December 2008 through 11 December 2008

ER -