Sports Type Classification using Signature Heatmaps

Publikation: Forskning - peer reviewKonferenceartikel i proceeding

Abstrakt

Automatic classification of activities in a sports arena is important in order to analyse and optimise the use of the arenas. In this work we classify five sports types based only on occupancy heatmaps produced from position data. Due to privacy issues we use thermal imaging for detecting people and then calculate their positions on the court us- ing homography. Heatmaps are produced by summarising Gaussian distributions respresenting people over 10-minute periods. Before classification the heatmaps are projected to a low-dimensional discriminative space using the principle of Fisherfaces. Our result using two weeks of video are very promising with a correct classification of 90.76 %.
Luk

Detaljer

Automatic classification of activities in a sports arena is important in order to analyse and optimise the use of the arenas. In this work we classify five sports types based only on occupancy heatmaps produced from position data. Due to privacy issues we use thermal imaging for detecting people and then calculate their positions on the court us- ing homography. Heatmaps are produced by summarising Gaussian distributions respresenting people over 10-minute periods. Before classification the heatmaps are projected to a low-dimensional discriminative space using the principle of Fisherfaces. Our result using two weeks of video are very promising with a correct classification of 90.76 %.
OriginalsprogEngelsk
TitelProceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Antal sider6
ForlagIEEE
Publikationsdato2013
Sider999-1004
ISBN (Trykt)9781479909940
ISBN (Elektronisk)978-0-7695-4990-3
DOI
StatusUdgivet - 2013
PublikationsartForskning
Peer reviewJa
Begivenhed2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) - Portland, Oregon, USA
Varighed: 23 jun. 201328 jun. 2013

Konference

Konference2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
LandUSA
ByPortland, Oregon
Periode23/06/201328/06/2013
SerieI E E E Conference on Computer Vision and Pattern Recognition. Proceedings
ISSN1063-6919

Download-statistik

Ingen data tilgængelig
ID: 75923648