A global perspective on MAP inference for low level vision

O. Woodford, C. Rother, V. Kolmogorov, in:, IEEE, 2009, pp. 2319–2326.

Download
No fulltext has been uploaded. References only!

Conference Paper | Published
Author
; ;
Abstract
In recent years the Markov Random Field (MRF) has become the de facto probabilistic model for low-level vision applications. However, in a maximum a posteriori (MAP) framework, MRFs inherently encourage delta function marginal statistics. By contrast, many low-level vision problems have heavy tailed marginal statistics, making the MRF model unsuitable. In this paper we introduce a more general Marginal Probability Field (MPF), of which the MRF is a special, linear case, and show that convex energy MPFs can be used to encourage arbitrary marginal statistics. We introduce a flexible, extensible framework for effectively optimizing the resulting NP-hard MAP problem, based around dual-decomposition and a modified mincost flow algorithm, and which achieves global optimality in some instances. We use a range of applications, including image denoising and texture synthesis, to demonstrate the benefits of this class of MPF over MRFs.
Publishing Year
Date Published
2009-05-01
Page
2319 - 2326
Conference
ICCV: International Conference on Computer Vision
IST-REx-ID

Cite this

Woodford O, Rother C, Kolmogorov V. A global perspective on MAP inference for low level vision. In: IEEE; 2009:2319-2326. doi:10.1109/ICCV.2009.5459434
Woodford, O., Rother, C., & Kolmogorov, V. (2009). A global perspective on MAP inference for low level vision (pp. 2319–2326). Presented at the ICCV: International Conference on Computer Vision, IEEE. https://doi.org/10.1109/ICCV.2009.5459434
Woodford, Oliver, Carsten Rother, and Vladimir Kolmogorov. “A Global Perspective on MAP Inference for Low Level Vision,” 2319–26. IEEE, 2009. https://doi.org/10.1109/ICCV.2009.5459434.
O. Woodford, C. Rother, and V. Kolmogorov, “A global perspective on MAP inference for low level vision,” presented at the ICCV: International Conference on Computer Vision, 2009, pp. 2319–2326.
Woodford O, Rother C, Kolmogorov V. 2009. A global perspective on MAP inference for low level vision. ICCV: International Conference on Computer Vision 2319–2326.
Woodford, Oliver, et al. A Global Perspective on MAP Inference for Low Level Vision. IEEE, 2009, pp. 2319–26, doi:10.1109/ICCV.2009.5459434.

Export

Marked Publications

Open Data IST Research Explorer

Search this title in

Google Scholar