Title: Graph Cut Algorithms for Binocular Stereo with Occlusions
1Graph Cut Algorithms for Binocular Stereo with
Occlusions
Vladimir Kolmogorov, Ramin Zabih
2Overview
- Traditional Stereo Methods
- Energy Minimization via Graph Cuts
- Stereo with Occlusions
- Voxel Labeling Algorithm
- Pixel Labeling Algorithm
- Results and Conclusions
3Traditional Stereo Methods
Traditional Stereo Problem
- pixel correspondences ? labeling (disparity)
4Traditional Stereo Methods Disparity
disparity
depth
? disparity depth
ground truth disparity
5Traditional Stereo MethodsBinocular Stereo
- goal is to compute pixels correspondences
- traditional stereo problem ? pixel labeling
problem - advantage can be solved by graph cuts
- problem is formulated as energy term
- new goal find the minimizing labeling
6Traditional Stereo MethodsEnergy Function
find labeling
that minimizes
cost for assigning labels
smoothness term
7Traditional Stereo MethodsEnergy Function
- data cost gives penalty for different
intensities
- smoothness term gives penalty for
discontinuities (Potts model)
other models absolute distance quadratic
8Energy Minimization via Graph Cuts
Max-flow / Min-Cut (Ford and Fulkerson Algorithm,
Push-Relabel Method)
9Energy Minimization via Graph Cuts
- convex V vs. metric / semimetric
- a-ß-swap move
- a-expansion move assigning label a to an
arbitrary set of pixels
10Stereo with Occlusions
11Stereo with Occlusions
- treat input symmetrically
- scene elements only visible in single view
- physically correct scenes ?
geometric constraints ? occlusions ?
physically possible labelings - introduce constraints in the problem formulation
- graph cuts perform unconstrained energy
minimization
12Voxel Labeling Algorithm
- discrete scene of voxels
- voxel v is active when visible from both cameras
- uniqueness constraint 11 correspondence of
pixels
13Voxel Labeling AlgorithmEnergy Function
smoothness term (Potts model)
matching penalty (only active voxels)
occlusion penalty
set of occluded pixels
14Pixel Labeling AlgorithmEnergy Function
- like traditional stereo but for both images e.g.
Potts model
active ?
15Minimizing the Energy
(0valid, else 8) uniqueness
- convert constrained into unconstrained
minimization problem - write as sum over pairs
- form of energy function standard stereo problem
- minimization with a-expansion algorithm
- modified definition of a-expansion move for voxel
labeling
16Results and Conclusions
ground truth
Tsukuba ref. image
17Results and Conclusions
- efficient energy minimization polynominal
time instead of exponential time - traditional stereo algorithm is faster
- pixel labeling better than voxel labeling
- prohibits holes in the scene
- allows to use other effective smoothness terms
- algorithms can be extended for multiple cameras
18Multi-view Stereo via Volumetric Graph Cuts
19Recent Work
- Graph-cut-based stereo matching using image
segmentation with symmetrical treatment of
occlusions, 2006 TUW
20Questions?
21References
- M. Bleyer, M. Gelautz, Graph-cut-based stereo
matching using image segmentation with
symmetrical treatment of occlusions, 2007 - Y. Boykov, O. Veksler, R. Zabih, Fast
Approximate Energy Minimization via Graph Cuts,
2001 - V. Kolmogorov, R. Zabih, Graph Cut Algorithms
for Binocular Stereo with Occlusions, 2005 - V. Kolmogorov, R. Zabih, What energy functions
can be minimized via graph cuts, 2004 - V. Kolmogorov, R. Zabih, Generalized
multi-camera scene reconstruction using graph
cuts, July 2003 - V. Kolmogorov, R. Zabih, Multi-camera Scene
Reconstruction via Graph Cuts, 2002 - S. Seits, C. Dyer, Photorealistic Scene
Reconstruction by Voxel Coloring, 1997 - R.Szeliski, R. Zabih, An Experimental Comparison
of Stereo Algorithms, 1999