Слайд 1 - Center For Machine Perception (Cmp)
Download
Report
Transcript Слайд 1 - Center For Machine Perception (Cmp)
A Lower Bound by One-against-all Decomposition for Potts
Model Energy Minimization
Alexander Shekhovtsov and Václav Hlaváč
Czech Technical University in Prague
Faculty of Electrical Engineering, Department of Cybernetics
Center for Machine Perception
Czech Republic
[email protected], [email protected]
Moravske Toplice, 2008
Motivation I
2/9
Energy Minimization Problem
Denoision, Boykov01
Stereo, Boykov01
Segmentation, Kovtun03
NP-hard; Many algorithms (Schlesinger76, Pearl88, Boykov01,
Wainwright03, Kolmogorov05, Komodakis05, Schlesinger07).
Algorithms LP-relaxation; Suboptimal LP solvers.
A faster LP solver for Potts model?
Alexander Shekhovtsov & Vaclav Hlavac, Prague 2008
Motivation II
3/9
Potts Model
Minimize the number
of steps
NP-hard for 3 labels
For 2 labels, it is solvable exactly by a min-cut / max-flow algorithm.
A natural heuristic: solve only 2 label problems
?
Alexander Shekhovtsov & Vaclav Hlavac, Prague 2008
?
Motivation II
4/9
Heuristic of Fangfang Lu et al. ACCV 2007
Fix labels in the areas
where labeling is
consistent
Is not guaranteed to
be correct
Can we propose a method which would fix provably optimal labels?
Alexander Shekhovtsov & Vaclav Hlavac, Prague 2008
Decompositions
5/9
Idea of decompositions by Wainwright03 (trees).
We propose a new kind of decomposition (one-against-all):
=
+
+
is equivalent to a binary problem (2 labels).
Solvable exactly by a single graph cut.
Alexander Shekhovtsov & Vaclav Hlavac, Prague 2008
Lower Bounds
6/9
The decomposition is not unique:
=
Free variables:
+
+
,
Theorem. Problem (*) is equivalent to standard LP-relaxation.
Coordinate ascent algorithm for (*).
Alexander Shekhovtsov & Vaclav Hlavac, Prague 2008
Per-node Bounds
7/9
Fix a node
Comput
e
If
- not optimal.
We obtain bounds of this type for free in our algorithm
If only one label remains in a pixel then we say that it is a part of any
optimal solution.
Alexander Shekhovtsov & Vaclav Hlavac, Prague 2008
Per-node Bounds: Experiments
8/9
Sample random problems: 10 x 10 grid graph with 5 labels.
Compute
, plot empirical estimate of
Problem parameters are sampled uniformly – almost no evidence for
optimal choice.
Real problems should be easier.
Alexander Shekhovtsov & Vaclav Hlavac, Prague 2008
Discussion
9/9
Our algorithm gets stuck in suboptimal points (satisfy necessary
conditions only but not sufficient ones).
We don’t know if it is faster than other algorithms.
Testing on real problems has to be performed.
We tested a BnB solver based on our bounds. Small problems were
solved exactly. It is important to have the ground truth.
Thank You
Alexander Shekhovtsov & Vaclav Hlavac, Prague 2008