Optimizing pixel labels for computer vision applications

Details Optimizing pixel labels for computer vision applications Optimizing pixel labels for computer vision applications

2007-06-15

2011-10-18

Zarka, David (Department: 2624)

Image analysis

Image segmentation

Reexamination Certificate

active

08041114

ABSTRACT:
Computer vision applications often require each pixel within an image to be assigned one of a set of labels. A method of improving the labels assigned to pixels is described which uses the quadratic pseudoboolean optimization (QPBO) algorithm. Starting with a partially labeled solution, an unlabeled pixel is assigned a value from a fully labeled reference solution and the energy of the partially labeled solution plus this additional pixel is calculated. The calculated energy is then used to generate a revised partially labeled solution using QPBO.

REFERENCES:
patent: 6078688 (2000-06-01), Cox et al.
patent: 6718063 (2004-04-01), Lennon et al.
patent: 6744923 (2004-06-01), Zabih et al.
patent: 6762769 (2004-07-01), Guo et al.
patent: 6973212 (2005-12-01), Boykov et al.
patent: 7139409 (2006-11-01), Paragios et al.
patent: 7209581 (2007-04-01), Zhang et al.
patent: 7212201 (2007-05-01), Geiger et al.
patent: 2005/0271273 (2005-12-01), Blake et al.
patent: 2006/0029275 (2006-02-01), Li et al.
patent: 2006/0039611 (2006-02-01), Rother et al.
patent: 2006/0104542 (2006-05-01), Blake et al.
patent: 2006/0147116 (2006-07-01), Le Clerc et al.
patent: 2006/0214932 (2006-09-01), Grady et al.
patent: 2006/0285747 (2006-12-01), Blake et al.
patent: 2006/0291721 (2006-12-01), Torr et al.
patent: 60200379 (1985-10-01), None
patent: 2002230540 (2002-08-01), None
patent: 2003222074 (2003-08-01), None
patent: 2007105368 (2007-04-01), None
patent: 2007192130 (2007-08-01), None
Rother et al., Optimizing Binary MRFs via Extended Roof Duality, cvpr, pp. 1-8, 2007 IEEE Conference on Computer Vision and Pattern Recognition, Jun. 17-22, 2007.
Kolmogorov et al., Minimizing non-submodular functions with graph cuts—a review, Microsoft Research Technical Report MSR-TR-2006-100, Jul. 2006, pp. 1-15.
Kolmogorov et al., Minimizing Nonsubmodular Functions with Graph Cuts—A Review, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29 , Issue 7, Jul. 2007, pp. 1274-1279.
Raj et al., MRF's for MRI's: Bayesian Reconstruction of MR Images via Graph Cuts, cvpr, vol. 1, pp. 1061-1068, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition—vol. 1 (CVPR'06), 2006.
Beasley, “Heuristic Algorithms for the Unconstrained Binary Quadratic Programming Problem”, 1998, pp. 36.
Soros, et al., “Local Search Heuristics for Unconstrained Quadratic Binary Optimization”, Rutcor Research Report, 2005, pp. 39.
Boros, et al., “Preprocessing of Unconstrained Quadratic Binary Optimization”, Rutcor Research Report, 2006, pp. 58.
Boros, et al., “Pseudo-Boolean Optimization”, 2001, pp. 83.
Boykov, et al., “Markov Random Fields with Efficient Approximations”, available at least as early as May 16, 2007, at <<http://www.cs.cornell.edu/˜rdz/Papers/BVZ-cvpr98.pdf>>, pp. 8.
Hammer, et al., “Roof Duality, Complementation and Persistency in Quadratic 0-1 Optimization”, Mathematical Programming, 1984, pp. 121-155.
Szeliski, et al., “A Comparative Study of Energy Minimization Methods for Markov Random Fields”, pp. 17.
Wu, et al., “Cross Entropy: A New Solver for Markov Random Field Modeling and Applications to Medical Image Segmentation”, at <<http://www.cs.ust.hk/˜achung/miccai05—wu—chung.pdf>>, Springer-Verlag Berlin Heidelberg, 2005, pp. 9.
PCT Search Report dated Nov. 25, 2008 for corresponding PCT Application No. PCT/US2008/065969, 4 pages.
Kohli, et al., “Efficiently Solving Dynamic Markov Random Fields Using Graph Cuts”, Tenth IEEE International Conference on Computer Vision, Oct. 17-21, 2005, pp. 8.
Kolmogorov, et al., “Minimizing Non-Submodular Functions With Graph Cuts—A Review”, Microsoft Research Technical Report: MSR-TR-2006-100, Jul. 2006, pp. 1-15.
Kolmogorov, et al., “What Energy Functions Can be Minimized Via Graph Cuts?”, IEEE Transactions on Pattern Anaylsis and Machine Intelligence, Feb. 2004, vol. 26 No. 2) pp. 147-159.

Affiliated with

Also associated with

Rating

Optimizing pixel labels for computer vision applications does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Optimizing pixel labels for computer vision applications, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Optimizing pixel labels for computer vision applications will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-4263914