Separating appearance from deformation

Proceedings of the IEEE International Conference on Computer Vision

Volumn 2, Issue , 2001, Pages 288-294

  Jojic, N ^a   Simard, P ^a   Frey, B J ^a   Heckerman, D ^a  

^a MICROSOFT RESEARCH   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

KNOWLEDGE REPRESENTATION; MATHEMATICAL MODELS; PROBABILITY DISTRIBUTIONS; WAVELET TRANSFORMS;

FEEDFORWARD DISCRIMINATORS; HANDWRITTEN DIGIT RECOGNITION;

IMAGE SEGMENTATION;

EID: 0034844742     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (14)

1. Maximum likelihood from incomplete data via the EM algorithm
2. Estimating mixture models of images and inferring spatial transformations using the EM algorithm
3. The EM algorithm for mixtures of factor analyzers
4. Modeling the manifolds of images of handwritten digits
5. Topographic transformation as a discrete latent variable
6. Efficient pattern recognition using a new transformation distance
7. Multiresolution tangent distance for affine invariant classification
8. Separating style from content
9. Learning parameterized models of image motion
10. Efficient region tracking with parametric models of geometry and illumination
11. Optical flow estimation using wavelet motion model
12. Design and use of linear models for image motion analysis
13. Determining correspondences for statistical models of facial appearance
14. A bootstrapping algorithm for learning linear models of object classes