Paper Info
Title
Image segmentation by figure-ground composition into maximal cliques
Abstract
We propose a mid-level statistical model for image segmentation that composes multiple figure-ground hypotheses (FG) obtained by applying constraints at different locations and scales, into larger interpretations (tilings) of the entire image. Inference is cast as optimization over sets of maximal cliques sampled from a graph connecting all non-overlapping figure-ground segment hypotheses. Potential functions over cliques combine unary, Gestalt-based figure qualities, and pairwise compatibilities among spatially neighboring segments, constrained by T-junctions and the boundary interface statistics of real scenes. Learning the model parameters is based on maximum likelihood, alternating between sampling image tilings and optimizing their potential function parameters. State of the art results are reported on the Berkeley and Stanford segmentation datasets, as well as VOC2009, where a 28% improvement was achieved.
Year
DOI
Venue
2011
10.1109/ICCV.2011.6126486
ICCV
Keywords
Field
DocType
stanford segmentation datasets,entire image,potential function parameter,figure-ground composition,potential function,sampling image tilings,model parameter,non-overlapping figure-ground segment hypothesis,maximal clique,mid-level statistical model,multiple figure-ground hypothesis,image segmentation,optimization,statistical model,edge detection,computer model,maximum likelihood,graph theory,graph connectivity,computational modeling,maximum likelihood estimation
Graph theory,Computer vision,Pairwise comparison,Scale-space segmentation,Pattern recognition,Unary operation,Computer science,Segmentation,Figure–ground,Image segmentation,Statistical model,Artificial intelligence
Conference
Volume
Issue
ISSN
2011
1
1550-5499
Citations 
PageRank 
References 
35
2.51
19
Authors
3
Name
Order
Citations
PageRank
Adrian Ion122221.11
João Carreira21680108.75
Cristian Sminchisescu33700244.41