LLE Algorithm in Natural Image Matting

Junbin Gao, David Tien, James Tulip

Research output: Book chapter/Published conference paperConference paper

36 Downloads (Pure)

Abstract

Accurately extracting foreground objects in images and video has wide applications in digital photography. These kind of problems are referred to as image matting. The most recent work [1] in natural image matting relies on local smoothness assumptions about foreground and background colours on which a cost function has been established. The closed-form solution has been derived based on a certain degree of user inputs. In this paper, we present a framework for formulating a new cost function from the manifold learning perspective based on the socalled Locally Linear Embedding [2] where the local smoothness assumptions have been replaced by an implicit manifold structure defined in local colour spaces. We illustrate our new algorithm using the standard benchmark images and very comparable results have been obtained.
Original languageEnglish
Title of host publicationProceedings of the The 7th International Conference on Information Technology and Applications
Subtitle of host publicationICITA 2011
Place of PublicationAustralia
PublisherICITA
Pages308-312
Number of pages5
Publication statusPublished - 2011
EventThe 7th International Conference on Information Technology and Applications: ICITA 2011 - Hilton Hotel, Sydney, Australia
Duration: 21 Nov 201124 Nov 2011
http://www.ieee.org/conferences_events/conferences/conferencedetails/index.html?Conf_ID=19233

Conference

ConferenceThe 7th International Conference on Information Technology and Applications
CountryAustralia
CitySydney
Period21/11/1124/11/11
Internet address

Fingerprint Dive into the research topics of 'LLE Algorithm in Natural Image Matting'. Together they form a unique fingerprint.

  • Cite this

    Gao, J., Tien, D., & Tulip, J. (2011). LLE Algorithm in Natural Image Matting. In Proceedings of the The 7th International Conference on Information Technology and Applications: ICITA 2011 (pp. 308-312). ICITA.