Face recognition using discriminatively trained orthogonal...

Image analysis – Applications – Personnel identification

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C382S155000, C382S159000, C382S181000, C382S190000

Reexamination Certificate

active

07936906

ABSTRACT:
Systems and methods are described for face recognition using discriminatively trained orthogonal rank one tensor projections. In an exemplary system, images are treated as tensors, rather than as conventional vectors of pixels. During runtime, the system designs visual features—embodied as tensor projections—that minimize intraclass differences between instances of the same face while maximizing interclass differences between the face and faces of different people. Tensor projections are pursued sequentially over a training set of images and take the form of a rank one tensor, i.e., the outer product of a set of vectors. An exemplary technique ensures that the tensor projections are orthogonal to one another, thereby increasing ability to generalize and discriminate image features over conventional techniques. Orthogonality among tensor projections is maintained by iteratively solving an ortho-constrained eigenvalue problem in one dimension of a tensor while solving unconstrained eigenvalue problems in additional dimensions of the tensor.

REFERENCES:
patent: 4790024 (1988-12-01), Peppers et al.
patent: 5754681 (1998-05-01), Watanabe et al.
patent: 5901244 (1999-05-01), Souma et al.
patent: 6441821 (2002-08-01), Nagasawa
patent: 6876755 (2005-04-01), Taylor et al.
patent: 7027620 (2006-04-01), Martinez
patent: 7715597 (2010-05-01), Costache et al.
patent: 2005/0105779 (2005-05-01), Kamei
patent: 2005/0123202 (2005-06-01), Hwang et al.
patent: 2006/0034517 (2006-02-01), Bober et al.
patent: 2007/0177805 (2007-08-01), Gallagher
patent: 2008/0014563 (2008-01-01), Visani et al.
patent: 2009/0074259 (2009-03-01), Baltatu et al.
patent: WO03088131 (2003-10-01), None
Hwann-Tzong Chen, Tyng-Luh Liu, Chiou-Shann Fuh, “Learning Effective Image Metrics from Few Pairwise Examples,” iccv, vol. 2, pp. 1371-1378, Tenth IEEE International Conference on Computer Vision (ICCV'05) vol. 2, Oct. 2005.
Dong Xu; Lin, S.; Shuicheng Yan; Xiaoou Tang; University of Science and Technology of China “Rank-one Projections with Adaptive Margins for Face Recognition” Computer Vision and Pattern Recognition, 2006 IEEE Computer Society Conference on, ISSN: 1063-6919 , Jun. 17-22, 2006.
Chopra, et al., “Learning a Similarity Metric Discriminatively, with Application to Face Verification”, available at least as early as Apr. 23, 2007, at <<http://www.cs.toronto.edu/˜hinton/csc2535/readings/chopra-05.pdf>>, pp. 1-8.
Shashua, et al., “Linear Image Coding for Regression and Classification using the Tensor-rank Principle”, available at least as early as Apr. 23, 2007, at <<http://www.cs.huji.ac.il/˜alevin/papers/rank1-cvpr01.ps>>, pp. 1-8.
Xu, et al., “Rank-one Projections with Adaptive Margins for Face Recognition”, retrieved on Apr. 24, 2007, at <<http://csdl2.computer.org/persagen/DLAbsToc.jsp?resourcePath=/dl/proceedings/&toc=comp/proceedings/cvpr/2006/2597/01/2597toc.xml&DOI=10.1109/CVPR.2006.234>>, IEEE, vol. 1, 2006, pp. 1-2.
Yan, et al., “Discriminant Analysis with Tensor Representation”, available at least as early as Apr. 23, 2007, at <<http://research.microsoft.com/users/leizhang/Paper/CVPR05-Shuicheng-Discriminant.pdf>>, IEEE, 2005, pp. 1-7.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Face recognition using discriminatively trained orthogonal... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Face recognition using discriminatively trained orthogonal..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Face recognition using discriminatively trained orthogonal... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2695329

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.