Background Subtraction for Time of Flight Imaging

A time of flight camera provides two types of images simultaneously, depth and intensity. In this paper a computational method for background subtraction, combining both images or fast sequences of images, is proposed. The background model is based on unbalanced or semi-supervised classifiers, in pa...

Descripción completa

Detalles Bibliográficos
Autores: Giacomantone, Javier, Violini, María Lucía, Lorenti, Luciano
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2017
País:Argentina
Institución:Universidad Nacional de La Plata
Repositorio:SEDICI (UNLP)
Idioma:inglés
OAI Identifier:oai:sedici.unlp.edu.ar:10915/62941
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/62941
Access Level:acceso abierto
Palabra clave:Ciencias Informáticas
industrial TOF cameras
machine vision
pattern recognition
support vector machines
Descripción
Sumario:A time of flight camera provides two types of images simultaneously, depth and intensity. In this paper a computational method for background subtraction, combining both images or fast sequences of images, is proposed. The background model is based on unbalanced or semi-supervised classifiers, in particular support vector machines. A brief review of one class support vector machines is first given. A method that combines the range and intensity data in two operational modes is then provided. Finally, experimental results are presented and discussed.