Learning to Segment 1st edition by Eran
Borenstein, Shimon Ullman ISBN 3540219828
9783540219828 pdf download
https://ebookball.com/product/learning-to-segment-1st-edition-by-
eran-borenstein-shimon-ullman-
isbn-3540219828-9783540219828-10662/
Explore and download more ebooks or textbooks
at ebookball.com
, Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats
A Constrained Semi supervised Learning Approach to Data Association
1st edition by Hendrik Kuck, Peter Carbonetto, Nando de Freitas ISBN
3540219828 9783540219828
https://ebookball.com/product/a-constrained-semi-supervised-
learning-approach-to-data-association-1st-edition-by-hendrik-
kuck-peter-carbonetto-nando-de-freitas-
isbn-3540219828-9783540219828-9412/
View Invariant Recognition Using Corresponding Object Fragments 1st
edition by Evgeniy Bart, Evgeny Byvatov, Shimon Ullman ISBN 3540219835
9783540219835
https://ebookball.com/product/view-invariant-recognition-using-
corresponding-object-fragments-1st-edition-by-evgeniy-bart-
evgeny-byvatov-shimon-ullman-isbn-3540219835-9783540219835-10540/
Bias in Shape Estimation 1st edition by Hui Ji, Cornelia Fermuller
ISBN 3540219828 9783540219828
https://ebookball.com/product/bias-in-shape-estimation-1st-
edition-by-hui-ji-cornelia-fermuller-
isbn-3540219828-9783540219828-9210/
Optimal Importance Sampling for Tracking in Image Sequences:
Application to Point Tracking 1st edition by Elise Arnaud, Etienne
Memin ISBN 3540219828 9783540219828
https://ebookball.com/product/optimal-importance-sampling-for-
tracking-in-image-sequences-application-to-point-tracking-1st-
edition-by-elise-arnaud-etienne-memin-
isbn-3540219828-9783540219828-10758/
,Intrinsic Images by Entropy Minimization 1st edition by Graham
Finlayson, Mark Drew, Cheng Lu ISBN 3540219828 9783540219828
https://ebookball.com/product/intrinsic-images-by-entropy-
minimization-1st-edition-by-graham-finlayson-mark-drew-cheng-lu-
isbn-3540219828-9783540219828-9776/
An Adaptive Window Approach for Image Smoothing and Structures
Preserving 1st edition by Charles Kervrann ISBN 3540219828
9783540219828
https://ebookball.com/product/an-adaptive-window-approach-for-
image-smoothing-and-structures-preserving-1st-edition-by-charles-
kervrann-isbn-3540219828-9783540219828-12036/
Combining Geometric and View-Based Approaches for Articulated Pose
Estimation 1st edition by David Demirdjian ISBN 3540219828
9783540219828
https://ebookball.com/product/combining-geometric-and-view-based-
approaches-for-articulated-pose-estimation-1st-edition-by-david-
demirdjian-isbn-3540219828-9783540219828-11362/
Tracking Articulated Motion Using a Mixture of Autoregressive Models
1st edition by Ankur Agarwal, Bill Triggs ISBN 3540219828
9783540219828
https://ebookball.com/product/tracking-articulated-motion-using-
a-mixture-of-autoregressive-models-1st-edition-by-ankur-agarwal-
bill-triggs-isbn-3540219828-9783540219828-13696/
An Information Based Measure for Grouping Quality 1st edition by Erik
Engbers, Michael Lindenbaum, Arnold Smeulders ISBN 3540219828
9783540219828
https://ebookball.com/product/an-information-based-measure-for-
grouping-quality-1st-edition-by-erik-engbers-michael-lindenbaum-
arnold-smeulders-isbn-3540219828-9783540219828-10624/
, Learning to Segment
Eran Borenstein and Shimon Ullman
Faculty of Mathematics and Computer Science
Weizmann Institute of Science
Rehovot, Israel 76100
{eran.borenstein,shimon.ullman}@weizmann.ac.il
Abstract. We describe a new approach for learning to perform class-
based segmentation using only unsegmented training examples. As in
previous methods, we first use training images to extract fragments that
contain common object parts. We then show how these parts can be
segmented into their figure and ground regions in an automatic learning
process. This is in contrast with previous approaches, which required
complete manual segmentation of the objects in the training examples.
The figure-ground learning combines top-down and bottom-up processes
and proceeds in two stages, an initial approximation followed by iterative
refinement. The initial approximation produces figure-ground labeling of
individual image fragments using the unsegmented training images. It
is based on the fact that on average, points inside the object are cov-
ered by more fragments than points outside it. The initial labeling is
then improved by an iterative refinement process, which converges in
up to three steps. At each step, the figure-ground labeling of individual
fragments produces a segmentation of complete objects in the training
images, which in turn induce a refined figure-ground labeling of the in-
dividual fragments. In this manner, we obtain a scheme that starts from
unsegmented training images, learns the figure-ground labeling of image
fragments, and then uses this labeling to segment novel images. Our ex-
periments demonstrate that the learned segmentation achieves the same
level of accuracy as methods using manual segmentation of training im-
ages, producing an automatic and robust top-down segmentation.
1 Introduction
The goal of figure-ground segmentation is to identify an object in the image and
separate it from the background. One approach to segmentation – the bottom-up
approach – is to first segment the image into regions and then identify the image
regions that correspond to a single object. The initial segmentation mainly relies
on image-based criteria, such as the grey level or texture uniformity of image
regions, as well as the smoothness and continuity of bounding contours. One
of the major shortcomings of the bottom-up approach is that an object may be
segmented into multiple regions, some of which may incorrectly merge the object
This research was supported in part by the Moross Laboratory at the Weizmann
Institute of Science.
T. Pajdla and J. Matas (Eds.): ECCV 2004, LNCS 3023, pp. 315–328, 2004.
c Springer-Verlag Berlin Heidelberg 2004
Borenstein, Shimon Ullman ISBN 3540219828
9783540219828 pdf download
https://ebookball.com/product/learning-to-segment-1st-edition-by-
eran-borenstein-shimon-ullman-
isbn-3540219828-9783540219828-10662/
Explore and download more ebooks or textbooks
at ebookball.com
, Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats
A Constrained Semi supervised Learning Approach to Data Association
1st edition by Hendrik Kuck, Peter Carbonetto, Nando de Freitas ISBN
3540219828 9783540219828
https://ebookball.com/product/a-constrained-semi-supervised-
learning-approach-to-data-association-1st-edition-by-hendrik-
kuck-peter-carbonetto-nando-de-freitas-
isbn-3540219828-9783540219828-9412/
View Invariant Recognition Using Corresponding Object Fragments 1st
edition by Evgeniy Bart, Evgeny Byvatov, Shimon Ullman ISBN 3540219835
9783540219835
https://ebookball.com/product/view-invariant-recognition-using-
corresponding-object-fragments-1st-edition-by-evgeniy-bart-
evgeny-byvatov-shimon-ullman-isbn-3540219835-9783540219835-10540/
Bias in Shape Estimation 1st edition by Hui Ji, Cornelia Fermuller
ISBN 3540219828 9783540219828
https://ebookball.com/product/bias-in-shape-estimation-1st-
edition-by-hui-ji-cornelia-fermuller-
isbn-3540219828-9783540219828-9210/
Optimal Importance Sampling for Tracking in Image Sequences:
Application to Point Tracking 1st edition by Elise Arnaud, Etienne
Memin ISBN 3540219828 9783540219828
https://ebookball.com/product/optimal-importance-sampling-for-
tracking-in-image-sequences-application-to-point-tracking-1st-
edition-by-elise-arnaud-etienne-memin-
isbn-3540219828-9783540219828-10758/
,Intrinsic Images by Entropy Minimization 1st edition by Graham
Finlayson, Mark Drew, Cheng Lu ISBN 3540219828 9783540219828
https://ebookball.com/product/intrinsic-images-by-entropy-
minimization-1st-edition-by-graham-finlayson-mark-drew-cheng-lu-
isbn-3540219828-9783540219828-9776/
An Adaptive Window Approach for Image Smoothing and Structures
Preserving 1st edition by Charles Kervrann ISBN 3540219828
9783540219828
https://ebookball.com/product/an-adaptive-window-approach-for-
image-smoothing-and-structures-preserving-1st-edition-by-charles-
kervrann-isbn-3540219828-9783540219828-12036/
Combining Geometric and View-Based Approaches for Articulated Pose
Estimation 1st edition by David Demirdjian ISBN 3540219828
9783540219828
https://ebookball.com/product/combining-geometric-and-view-based-
approaches-for-articulated-pose-estimation-1st-edition-by-david-
demirdjian-isbn-3540219828-9783540219828-11362/
Tracking Articulated Motion Using a Mixture of Autoregressive Models
1st edition by Ankur Agarwal, Bill Triggs ISBN 3540219828
9783540219828
https://ebookball.com/product/tracking-articulated-motion-using-
a-mixture-of-autoregressive-models-1st-edition-by-ankur-agarwal-
bill-triggs-isbn-3540219828-9783540219828-13696/
An Information Based Measure for Grouping Quality 1st edition by Erik
Engbers, Michael Lindenbaum, Arnold Smeulders ISBN 3540219828
9783540219828
https://ebookball.com/product/an-information-based-measure-for-
grouping-quality-1st-edition-by-erik-engbers-michael-lindenbaum-
arnold-smeulders-isbn-3540219828-9783540219828-10624/
, Learning to Segment
Eran Borenstein and Shimon Ullman
Faculty of Mathematics and Computer Science
Weizmann Institute of Science
Rehovot, Israel 76100
{eran.borenstein,shimon.ullman}@weizmann.ac.il
Abstract. We describe a new approach for learning to perform class-
based segmentation using only unsegmented training examples. As in
previous methods, we first use training images to extract fragments that
contain common object parts. We then show how these parts can be
segmented into their figure and ground regions in an automatic learning
process. This is in contrast with previous approaches, which required
complete manual segmentation of the objects in the training examples.
The figure-ground learning combines top-down and bottom-up processes
and proceeds in two stages, an initial approximation followed by iterative
refinement. The initial approximation produces figure-ground labeling of
individual image fragments using the unsegmented training images. It
is based on the fact that on average, points inside the object are cov-
ered by more fragments than points outside it. The initial labeling is
then improved by an iterative refinement process, which converges in
up to three steps. At each step, the figure-ground labeling of individual
fragments produces a segmentation of complete objects in the training
images, which in turn induce a refined figure-ground labeling of the in-
dividual fragments. In this manner, we obtain a scheme that starts from
unsegmented training images, learns the figure-ground labeling of image
fragments, and then uses this labeling to segment novel images. Our ex-
periments demonstrate that the learned segmentation achieves the same
level of accuracy as methods using manual segmentation of training im-
ages, producing an automatic and robust top-down segmentation.
1 Introduction
The goal of figure-ground segmentation is to identify an object in the image and
separate it from the background. One approach to segmentation – the bottom-up
approach – is to first segment the image into regions and then identify the image
regions that correspond to a single object. The initial segmentation mainly relies
on image-based criteria, such as the grey level or texture uniformity of image
regions, as well as the smoothness and continuity of bounding contours. One
of the major shortcomings of the bottom-up approach is that an object may be
segmented into multiple regions, some of which may incorrectly merge the object
This research was supported in part by the Moross Laboratory at the Weizmann
Institute of Science.
T. Pajdla and J. Matas (Eds.): ECCV 2004, LNCS 3023, pp. 315–328, 2004.
c Springer-Verlag Berlin Heidelberg 2004
