GPU based GMM segmentation of kinect data

Author

Amamra, Abdenour ; Mouats, Tarek ; Aouf, Nabil

Author_Institution

Dept. of Inf. & Syst. Eng., Cranfield Univ., Shrivenham, UK

fYear

2014

fDate

10-12 Sept. 2014

Firstpage

1

Lastpage

4

Abstract

This paper presents a novel approach for background/foreground segmentation of RGBD data with the Gaussian Mixture Models (GMM). We first start by the background subtraction from the colour and depth images separately. The foregrounds resulting from both streams are then fused for a more accurate detection. Our segmentation solution is implemented on the GPU. Thus, it works at the full frame rate of the sensor (30fps). Test results show its robustness against illumination change, shadows and reflections.

Keywords

Gaussian processes; graphics processing units; image colour analysis; image segmentation; image sensors; mixture models; GPU-based GMM segmentation; Gaussian mixture models; Kinect data; RGBD data; RGBD sensors; background segmentation; background subtraction; colour images; depth images; foreground segmentation; Arrays; Graphics processing units; Image color analysis; Image segmentation; Lighting; Real-time systems; Robot sensing systems; Background substraction; Gaussian Mixture Models; RGBD sensors; image data fusion; real-time tracking;

fLanguage

English

Publisher

ieee

Conference_Titel

ELMAR (ELMAR), 2014 56th International Symposium

Conference_Location

Zadar

Type

conf

DOI

10.1109/ELMAR.2014.6923325

Filename

6923325