Near-Real-Time Image Matting with Known Background

Author

Gong, Minglun ; Yang, Yee-Hong

Author_Institution

Memorial Univ. of Newfoundland, St. John´´s, NL, Canada

fYear

2009

fDate

25-27 May 2009

Firstpage

81

Lastpage

87

Abstract

This paper presents a novel matting algorithm for handling cases where foreground objects are in front of complex (non-smooth) but known background. The algorithm is based on a Poisson equation that takes the gradient of the known background into consideration. The quantitative evaluation based on ground truth shows that the matting results obtained using the proposed algorithm is more accurate than the ones generated by the global Poisson matting. The proposed algorithm is also optimized for parallel processing and runs in near-real-time on programmable graphics hardware.

Keywords

Poisson equation; image colour analysis; image morphing; Poisson equation; complex background; foreground objects; global Poisson matting; matting algorithm; near-real-time image matting; parallel processing; programmable graphics hardware; Computer graphics; Computer vision; Hardware; Layout; Mobile computing; Parallel processing; Poisson equations; Portable computers; Robot vision systems; Video sequences; Image matting; Poisson equation; Programmable graphics hardware;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Robot Vision, 2009. CRV '09. Canadian Conference on

Conference_Location

Kelowna, BC

Print_ISBN

978-0-7695-3651-4

Type

conf

DOI

10.1109/CRV.2009.7

Filename

5230535