Evaluating Hierarchical Graph-based Segmentation

Using real world images, two hierarchical graph-based segmentation methods are evaluated with respect to segmentations produced by humans. Global and local consistency measures do not show big differences between the two representative methods although human visual inspection of the results show advantages for one method. To a certain extent this subjective impression is captured by the new criteria of 'region size variation'.

• Type: Conference
• Year: 2006
• Where: ICPR
• Authors: Yll Haxhimusa, Adrian Ion, Walter G. Kropatsch
• Keywords: Computer Vision, Graph-based Segmentation Methods, Human Visual Inspection, ICPR 2006, Local Consistency Measures