Paper Info
Title
Interactive image segmentation by matching attributed relational graphs
Abstract
A model-based graph matching approach is proposed for interactive image segmentation. It starts from an over-segmentation of the input image, exploiting color and spatial information among regions to propagate the labels from the regions marked by the user-provided seeds to the entire image. The region merging procedure is performed by matching two graphs: the input graph, representing the entire image; and the model graph, representing only the marked regions. The optimization is based on discrete search using deformed graphs to efficiently evaluate the spatial information. Note that by using a model-based approach, different interactive segmentation problems can be tackled: binary and multi-label segmentation of single images as well as of multiple similar images. Successful results for all these cases are presented, in addition to a comparison between our binary segmentation results and those obtained with state-of-the-art approaches. An implementation is available at http://structuralsegm.sourceforge.net/.
Year
DOI
Venue
2012
10.1016/j.patcog.2011.08.017
Pattern Recognition
Keywords
Field
DocType
multi-label segmentation,multiple similar image,entire image,different interactive segmentation problem,interactive image segmentation,input image,spatial information,binary segmentation result,relational graph,single image,input graph
Computer vision,Scale-space segmentation,Pattern recognition,Range segmentation,Image texture,Segmentation-based object categorization,Matching (graph theory),Image segmentation,Artificial intelligence,Connected-component labeling,Mathematics,Minimum spanning tree-based segmentation
Journal
Volume
Issue
ISSN
45
3
0031-3203
Citations 
PageRank 
References 
26
0.85
32
Authors
5
Name
Order
Citations
PageRank
Alexandre Noma1483.66
Ana B. V. Graciano2260.85
Roberto M. Cesar, Jr.379449.46
Luis A. Consularo4260.85
Isabelle Bloch52123170.75