Unsupervised co-segmentation through region matching

Abstract : Co-segmentation is defined as jointly partitioning multiple images depicting the same or similar object, into foreground and background. Our method consists of a multiple-scale multiple-image generative model, which jointly estimates the foreground and background appearance distributions from several images, in a non-supervised manner. In contrast to other co-segmentation methods, our approach does not require the images to have similar foregrounds and different backgrounds to function properly. Region matching is applied to exploit inter-image information by establishing correspondences between the common objects that appear in the scene. Moreover, computing many-to-many associations of regions allow further applications, like recognition of object parts across images. We report results on iCoseg, a challenging dataset that presents extreme variability in camera viewpoint, illumination and object deformations and poses. We also show that our method is robust against large intra-class variability in the MSRC database.
Document type :
Conference papers
Liste complète des métadonnées

Contributor : Enzo Ferrante <>
Submitted on : Friday, August 30, 2013 - 5:21:27 PM
Last modification on : Tuesday, February 5, 2019 - 1:52:14 PM


  • HAL Id : hal-00856291, version 1



Jose Rubio, Joan Serrat, Antonio López, Nikos Paragios. Unsupervised co-segmentation through region matching. 25th IEEE Conference on Computer Vision and Pattern Recognition - CVPR 2012, Jun 2012, Rhode Island, United States. pp.749-756. ⟨hal-00856291⟩



Record views