TY - GEN
T1 - Flattening supervoxel hierarchies by the uniform entropy slice
AU - Xu, Chenliang
AU - Whitt, Spencer
AU - Corso, Jason J.
PY - 2013
Y1 - 2013
N2 - Supervoxel hierarchies provide a rich multiscale decomposition of a given video suitable for subsequent processing in video analysis. The hierarchies are typically computed by an unsupervised process that is susceptible to under-segmentation at coarse levels and over-segmentation at fine levels, which make it a challenge to adopt the hierarchies for later use. In this paper, we propose the first method to overcome this limitation and flatten the hierarchy into a single segmentation. Our method, called the uniform entropy slice, seeks a selection of supervoxels that balances the relative level of information in the selected supervoxels based on some post hoc feature criterion such as object-ness. For example, with this criterion, in regions nearby objects, our method prefers finer supervoxels to capture the local details, but in regions away from any objects we prefer coarser supervoxels. We formulate the uniform entropy slice as a binary quadratic program and implement four different feature criteria, both unsupervised and supervised, to drive the flattening. Although we apply it only to supervoxel hierarchies in this paper, our method is generally applicable to segmentation tree hierarchies. Our experiments demonstrate both strong qualitative performance and superior quantitative performance to state of the art baselines on benchmark internet videos.
AB - Supervoxel hierarchies provide a rich multiscale decomposition of a given video suitable for subsequent processing in video analysis. The hierarchies are typically computed by an unsupervised process that is susceptible to under-segmentation at coarse levels and over-segmentation at fine levels, which make it a challenge to adopt the hierarchies for later use. In this paper, we propose the first method to overcome this limitation and flatten the hierarchy into a single segmentation. Our method, called the uniform entropy slice, seeks a selection of supervoxels that balances the relative level of information in the selected supervoxels based on some post hoc feature criterion such as object-ness. For example, with this criterion, in regions nearby objects, our method prefers finer supervoxels to capture the local details, but in regions away from any objects we prefer coarser supervoxels. We formulate the uniform entropy slice as a binary quadratic program and implement four different feature criteria, both unsupervised and supervised, to drive the flattening. Although we apply it only to supervoxel hierarchies in this paper, our method is generally applicable to segmentation tree hierarchies. Our experiments demonstrate both strong qualitative performance and superior quantitative performance to state of the art baselines on benchmark internet videos.
UR - http://www.scopus.com/inward/record.url?scp=84898832980&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84898832980&partnerID=8YFLogxK
U2 - 10.1109/ICCV.2013.279
DO - 10.1109/ICCV.2013.279
M3 - Conference contribution
AN - SCOPUS:84898832980
SN - 9781479928392
T3 - Proceedings of the IEEE International Conference on Computer Vision
SP - 2240
EP - 2247
BT - Proceedings - 2013 IEEE International Conference on Computer Vision, ICCV 2013
T2 - 2013 14th IEEE International Conference on Computer Vision, ICCV 2013
Y2 - 1 December 2013 through 8 December 2013
ER -