SCALPEL: Segmentation Cascades with Localized Priors and Efficient Learning
Authors:
- David WeissDWDavid Weiss
- Ben TaskarBTBen Taskar
- 1 June 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2035-2042
Abstract
We propose SCALPEL, a flexible method for object segmentation that integrates rich region-merging cues with mid- and high-level information about object layout, class, and scale into the segmentation process. Unlike competing approaches, SCALPEL uses a cascade of bottom-up segmentation models that is capable of learning to ignore boundaries early on, yet use them as a stopping criterion once the object has been mostly segmented. Furthermore, we show how such cascades can be learned efficiently. When paired with a novel method that generates better localized shape priors than our competitors, our method leads to a concise, accurate set of segmentation proposals, these proposals are more accurate on the PASCAL VOC2010 dataset than state-of-the-art methods that use re-ranking to filter much larger bags of proposals. The code for our algorithm is available online.Keywords
This publication has 14 references indexed in Scilit:
- Image Segmentation by Cascaded Region Agglomeration2013
- Figure-ground segmentation by transferring window masks2012
- Extracting foreground masks towards object recognition2011
- Ensemble of exemplar-SVMs for object detection and beyond2011
- Image segmentation by figure-ground composition into maximal cliques2011
- Segmentation as selective search for object recognition2011
- The truth about cats and dogs2011
- Learning a category independent object detection cascade2011
- Layered Object Models for Image SegmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
- Learning a classification model for segmentation2003