Sciweavers

Share
ECCV
2010
Springer

Superpixels and Supervoxels in an Energy Optimization Framework

3 years 8 months ago
Superpixels and Supervoxels in an Energy Optimization Framework
Many methods for object recognition, segmentation, etc., rely on tessellation of an image into "superpixels". A superpixel is an image patch which is better aligned with intensity edges than a rectangular patch. Superpixels can be extracted with any segmentation algorithm, however, most of them produce highly irregular superpixels, with widely varying sizes and shapes. A more regular space tessellation may be desired. We formulate the superpixel partitioning problem in an energy minimization framework, and optimize with graph cuts. Our energy function explicitly encourages regular superpixels. We explore variations of the basic energy, which allow a trade-off between a less regular tessellation but more accurate boundaries or better efficiency. Our advantage over previous work is computational efficiency, principled optimization, and applicability to 3D "supervoxel" segmentation. We achieve high boundary recall on 2D images and spatial coherence on video. We also sh...
Olga Veksler, Yuri Boykov, Paria Mehrani
Added 06 Dec 2010
Updated 06 Dec 2010
Type Conference
Year 2010
Where ECCV
Authors Olga Veksler, Yuri Boykov, Paria Mehrani
Comments (0)
books