Skip to content

view

Jiří Borovec
presents
Superpixel segmentation and region growing
 
On 2017-11-23 11:00 at G205
 
The purpose of this seminar is to present two of our recently accepted papers:

Borovec J., Švihlík J., Kybic J. and Habart D. "Supervised and unsupervised
segmentation using superpixels, model estimation, and Graph Cut." Journal of
Electronic Imaging, 2017.
ftp://cmp.felk.cvut.cz/pub/cmp/articles/borovec/Borovec-JEI2017-Segm.pdf

Borovec J., Kybic J. and Sugimoto A.. "Region growing using superpixels with
learned shape prior." Journal of Electronic Imaging, 2017.
ftp://cmp.felk.cvut.cz/pub/cmp/articles/borovec/Borovec-JEI2017-RG2Sp.pdf

The papers are related, that is why they are presented together.

Abstract for the first paper:

It is often advantageous to first group pixels into compact, edge-respecting
superpixels, because these reduce the size of the segmentation problem and thus
the segmentation time by an order of magnitudes.
In addition, features calculated from superpixel regions are more robust than
features calculated from fixed pixel neighbourhoods.
We present a~fast and general multi-class image segmentation method consisting
of the following steps:
 (i) computation of superpixels;
 (ii) extraction of superpixel-based descriptors;
 (iii) calculating image-based class probabilities in a~supervised or
unsupervised manner;
 and (iv) regularized superpixel classification using Graph~Cut.
We apply this segmentation pipeline to real-world medical imaging applications
and present obtained performance.
We show that unsupervised segmentation provides similar results to the
supervised method, but does not require manually annotated training data, which
is often expensive to obtain.

------------------------------------

Abstract for the second paper:

Region growing is a~classical image segmentation method based on hierarchical
region aggregation using local similarity rules. Our proposed method differs
from classical region growing in three important aspects. 
First, it works on the level of superpixels instead of pixels, which leads to
a~substantial speedup. 
Second, our method uses learned statistical shape properties that encourage
plausible shapes. In particular, we use ray features to describe the object
boundary. 
Third, our method can segment multiple objects and ensure that the
segmentations
do not overlap. The problem is represented as an energy minimization and is
solved either greedily, or iteratively using Graph~Cuts.
Back to the list