An In-depth Study of Sparse Codes on Abnormality Detection

Publikation: Forskning - peer reviewKonferenceartikel i proceeding

Abstrakt

Sparse representation has been applied successfully in
abnormal event detection, in which the baseline is to learn
a dictionary accompanied by sparse codes. While much
emphasis is put on discriminative dictionary construction,
there are no comparative studies of sparse codes regarding abnormality detection. We present an in-depth study of
two types of sparse codes solutions - greedy algorithms and
convex L1-norm solutions - and their impact on abnormal-
ity detection performance. We also propose our framework
of combining sparse codes with different detection methods.
Our comparative experiments are carried out from various angles to better understand the applicability of sparse
codes, including computation time, reconstruction error,
sparsity, detection accuracy, and their performance combining various detection methods. The experiment results
show that combining OMP codes with maximum coordinate
detection could achieve state-of-the-art performance on the
UCSD dataset [14].
Luk

Detaljer

Sparse representation has been applied successfully in
abnormal event detection, in which the baseline is to learn
a dictionary accompanied by sparse codes. While much
emphasis is put on discriminative dictionary construction,
there are no comparative studies of sparse codes regarding abnormality detection. We present an in-depth study of
two types of sparse codes solutions - greedy algorithms and
convex L1-norm solutions - and their impact on abnormal-
ity detection performance. We also propose our framework
of combining sparse codes with different detection methods.
Our comparative experiments are carried out from various angles to better understand the applicability of sparse
codes, including computation time, reconstruction error,
sparsity, detection accuracy, and their performance combining various detection methods. The experiment results
show that combining OMP codes with maximum coordinate
detection could achieve state-of-the-art performance on the
UCSD dataset [14].
OriginalsprogEngelsk
TitelIEEE Advanced Video and Signal-based Surveillance (AVSS) 2016
Antal sider7
UdgiverIEEE
Publikationsdato2016
Sider1-7
ISBN (trykt)978-1-5090-3812-1
ISBN (elektronisk)978-1-5090-3811-4
DOI
StatusUdgivet - 2016
Begivenhed - Colorado Springs, USA

Konference

KonferenceThe 13th International Conference on Advanced Video and Signal-Based Surveillance
Nummer13
LandUSA
By Colorado Springs
Periode23/08/201626/08/2016
Internetadresse

Download-statistik

Ingen data tilgængelig
ID: 238286204