A motion-based segmentation scheme for object-oriented video coding

Yuanqun Luo, Donglai Xu, Ian French

Research output: Contribution to conferencePaperpeer-review

Abstract

This paper presents a new video segmentation scheme for object-oriented MPEG-4 video coding. In the scheme, the object motion is modeled by quadratic transform with coefficients estimated via a nonlinear Quasi-Newton method. To effectively merge the regions with similar motions within an image, a region merging process, which is guided by a weighted region adjacency graph, is employed. Simulation results show the proposed scheme provides high performance in terms of segmentation accuracy and video compression ratio.
Original languageEnglish
Publication statusPublished - 2004
EventInternational workshop on systems, signals and image processing. - Poznan, Poland
Duration: 13 Sept 200415 Sept 2004

Conference

ConferenceInternational workshop on systems, signals and image processing.
Country/TerritoryPoland
CityPoznan
Period13/09/0415/09/04

Bibliographical note

This paper presents a new video segmentation scheme for object-oriented MPEG-4 video coding. In the scheme, the object motion is modeled by quadratic transform with coefficients estimated via a nonlinear Quasi-Newton method. To effectively merge the regions with similar motions within an image, a region merging process, which is guided by a weighted region adjacency graph, is employed. Simulation results show the proposed scheme provides high performance in terms of segmentation accuracy and video compression ratio.

Fingerprint

Dive into the research topics of 'A motion-based segmentation scheme for object-oriented video coding'. Together they form a unique fingerprint.

Cite this