Multi-view object co-segmentation based on the mixture of links model

Dongting Hu, Yugang Li, Xiabi Liu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We present a novel mixture of links model to segment an object observed from multiple viewpoints. Each component in this mixture represents a temporal linkage between superpixels from all the viewpoints, hence expressing the inter-view consistency. The principle goal is to find the maximum a posterior estimate of appearance models and the exact bounding-box of object in each view. To this end, the segmentation is casted as finding more comprehensive and accurate samples using the mixture of links model. In contrast to most existing multi-view co-segmentation methods that rely on time-consuming 3D information, our method only uses 2D cues to achieve faster speed without decreasing the accuracy. The experimental results confirm the effectiveness of our approach.

Original languageEnglish
Title of host publicationSeventh International Conference on Digital Image Processing, ICDIP 2015
EditorsCharles M. Falco, Xudong Jiang
PublisherSPIE
ISBN (Electronic)9781628418293
DOIs
Publication statusPublished - 2015
Event7th International Conference on Digital Image Processing, ICDIP 2015 - Los Angeles, United States
Duration: 9 Apr 201510 Apr 2015

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9631
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference7th International Conference on Digital Image Processing, ICDIP 2015
Country/TerritoryUnited States
CityLos Angeles
Period9/04/1510/04/15

Keywords

  • Image segmentation
  • Mixture of links model
  • Multi-view

Fingerprint

Dive into the research topics of 'Multi-view object co-segmentation based on the mixture of links model'. Together they form a unique fingerprint.

Cite this