Loading...
Please wait, while we are loading the content...
Similar Documents
Image segmentation by figure-ground composition into maximal cliques (2011)
| Content Provider | CiteSeerX |
|---|---|
| Author | Ion, Adrian Carreira, Joao Sminchisescu, Cristian |
| Description | 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. 1. In International Conference on Computer Vision |
| File Format | |
| Language | English |
| Publisher Date | 2011-01-01 |
| Access Restriction | Open |
| Subject Keyword | Entire Image Mid-level Statistical Model Image Tiling Image Segmentation Gestalt-based Figure Quality Maximum Likelihood Non-overlapping Figure-ground Segment Hypothesis Boundary Interface Statistic Potential Function Maximal Clique Figure-ground Composition Different Location Potential Function Parameter Model Parameter Stanford Segmentation Datasets Multiple Figure-ground Hypothesis Art Result Real Scene |
| Content Type | Text |
| Resource Type | Article |
