Spatio-temporal shape contexts for human action retrieval

Ling Shao; Yuanjia Du

doi:10.1145/1631040.1631048

Spatio-temporal shape contexts for human action retrieval

Ling Shao, Yuanjia Du

School of Computer Science

Research output: Contribution to conference › Paper

11 Citations (Scopus)

Abstract

A novel method using the combination of spatio-temporal interest points and spatio-temporal shape contexts for human action description and retrieval is proposed. Both the feature points detection and description are invariant to geometric and photometric changes and intra-class variations. The experimental results show that the proposed descriptor is more effective than brightness gradients based motion descriptors for classifying and retrieving challenging human actions.

Original language	English
DOIs	https://doi.org/10.1145/1631040.1631048
Publication status	Published - 2009
Event	IMCE '09 - 1st International Workshop on Interactive Multimedia for Consumer Electronics - Beijing, China Duration: 1 Oct 2009 → …

Workshop

Workshop	IMCE '09 - 1st International Workshop on Interactive Multimedia for Consumer Electronics
Period	1/10/09 → …

Keywords

Motion descriptor
spatio-temporal shape context
action retrieval

Access to Document

10.1145/1631040.1631048

Cite this

@conference{f0bf06bc3bdc42208b6a3e109da00f03,

title = "Spatio-temporal shape contexts for human action retrieval",

abstract = "A novel method using the combination of spatio-temporal interest points and spatio-temporal shape contexts for human action description and retrieval is proposed. Both the feature points detection and description are invariant to geometric and photometric changes and intra-class variations. The experimental results show that the proposed descriptor is more effective than brightness gradients based motion descriptors for classifying and retrieving challenging human actions.",

keywords = "Motion descriptor, spatio-temporal shape context, action retrieval",

author = "Ling Shao and Yuanjia Du",

year = "2009",

doi = "10.1145/1631040.1631048",

language = "English",

note = "IMCE '09 - 1st International Workshop on Interactive Multimedia for Consumer Electronics ; Conference date: 01-10-2009",

}

TY - CONF

T1 - Spatio-temporal shape contexts for human action retrieval

AU - Shao, Ling

AU - Du, Yuanjia

PY - 2009

Y1 - 2009

N2 - A novel method using the combination of spatio-temporal interest points and spatio-temporal shape contexts for human action description and retrieval is proposed. Both the feature points detection and description are invariant to geometric and photometric changes and intra-class variations. The experimental results show that the proposed descriptor is more effective than brightness gradients based motion descriptors for classifying and retrieving challenging human actions.

AB - A novel method using the combination of spatio-temporal interest points and spatio-temporal shape contexts for human action description and retrieval is proposed. Both the feature points detection and description are invariant to geometric and photometric changes and intra-class variations. The experimental results show that the proposed descriptor is more effective than brightness gradients based motion descriptors for classifying and retrieving challenging human actions.

KW - Motion descriptor

KW - spatio-temporal shape context

KW - action retrieval

UR - https://www.scopus.com/pages/publications/72049120990

U2 - 10.1145/1631040.1631048

DO - 10.1145/1631040.1631048

M3 - Paper

T2 - IMCE '09 - 1st International Workshop on Interactive Multimedia for Consumer Electronics

Y2 - 1 October 2009

ER -

Spatio-temporal shape contexts for human action retrieval

Abstract

Workshop

Keywords

Access to Document

Other files and links

Fingerprint

Cite this