GPGPU implementation of an improved nonparametric background modeling for moving object detection strategies

GPGPU implementation of an improved nonparametric background modeling for moving object detection strategies

A GPGPU-based real-time nonparametric modeling strategy for moving object detection is proposed. By dynamically estimating bandwidth matrices and by using a selective update mechanism, previous approaches are outperformed in quality while preserving their computational requirements.

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cuevas, C., Berjon, D., Moran, F., Garcia, N.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Bandwidth
Computation
Computational modeling
Computer vision
Conferences
Consumer electronics
Consumption
Electronics
Hardware
Object detection
Preserving
Real time
Real-time systems
Strategy
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A GPGPU-based real-time nonparametric modeling strategy for moving object detection is proposed. By dynamically estimating bandwidth matrices and by using a selective update mechanism, previous approaches are outperformed in quality while preserving their computational requirements.
ISSN:2158-3994
2158-4001
DOI:10.1109/ICCE.2013.6486778