%0 Generic %1 Pont-Tuset2012 %A Pont-Tuset, J %A Marqués, F %B IEEE ECCV %D 2012 %K imported %T Supervised Assessment of Segmentation Hierarchies %X <p>This paper addresses the problem of the supervised assessment of hierarchical region-based image representations. Given the large amount of partitions represented in such structures, the supervised assessment approaches in&nbsp;the literature are based on selecting a reduced set of representative partitions and&nbsp;evaluating their quality. Assessment results, therefore, depend on the partition selection strategy used. Instead, we propose to find the partition in the tree that best&nbsp;matches the ground-truth partition, that is, the upper-bound partition selection.</p><p>We show that different partition selection algorithms can lead to different conclusions regarding the quality of the assessed trees and that the <strong>upper-bound partition&nbsp;selection</strong> provides the following advantages: 1) it does not limit the assessment&nbsp;to a reduced set of partitions, and 2) it better discriminates the random trees from&nbsp;actual ones, which reflects a better qualitative behavior. We model the problem as&nbsp;a Linear Fractional Combinatorial Optimization (LFCO) problem, which makes&nbsp;the upper-bound selection feasible and efficient.</p>

@conference{Pont-Tuset2012, abstract = {<p>This paper addresses the problem of the supervised assessment of hierarchical region-based image representations. Given the large amount of partitions represented in such structures, the supervised assessment approaches in\&nbsp;the literature are based on selecting a reduced set of representative partitions and\&nbsp;evaluating their quality. Assessment results, therefore, depend on the partition selection strategy used. Instead, we propose to find the partition in the tree that best\&nbsp;matches the ground-truth partition, that is, the upper-bound partition selection.</p><p>We show that different partition selection algorithms can lead to different conclusions regarding the quality of the assessed trees and that the <strong>upper-bound partition\&nbsp;selection</strong> provides the following advantages: 1) it does not limit the assessment\&nbsp;to a reduced set of partitions, and 2) it better discriminates the random trees from\&nbsp;actual ones, which reflects a better qualitative behavior. We model the problem as\&nbsp;a Linear Fractional Combinatorial Optimization (LFCO) problem, which makes\&nbsp;the upper-bound selection feasible and efficient.</p>}, added-at = {2013-09-29T14:16:50.000+0200}, author = {Pont-Tuset, J and Marqu\'{e}s, F}, biburl = {https://www.bibsonomy.org/bibtex/2eeb33817b3f5478ee003b00ee57a5820/guillem.palou}, booktitle = {IEEE ECCV}, interhash = {e3ef93a894e7dc871af04481cb702673}, intrahash = {eeb33817b3f5478ee003b00ee57a5820}, keywords = {imported}, timestamp = {2013-09-29T14:16:50.000+0200}, title = {{Supervised Assessment of Segmentation Hierarchies}}, year = 2012 }