Learning topographic representation for multi-view image patterns

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Volumn 2, Issue , 2001, Pages 1329-1332

  Li, Stan Z ^a   Lv, Xiao Guang ^a   Zhang, Hong Jiang ^a   Fu, Qing Dong ^b   Cheng, Yimin ^b  

^a MICROSOFT RESEARCH ASIA   (China)

^b UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA   (China)

Author keywords

[No Author keywords available]

Indexed keywords

COMPUTER SIMULATION; COMPUTER SOFTWARE; FACE RECOGNITION; IMAGE ANALYSIS; IMAGE PROCESSING; INDEPENDENT COMPONENT ANALYSIS; LIGHTING; PRINCIPAL COMPONENT ANALYSIS; TWO DIMENSIONAL;

MULTI-VIEW IMAGE PATTERNS; TOPOGRAPHIC INDEPENDENT COMPONENT ANALYSIS;

LEARNING SYSTEMS;

EID: 0034843019     PISSN: 15206149     EISSN: None     Source Type: Journal    
DOI: 10.1109/ICASSP.2001.941171     Document Type: Article

References (19)

1. Parametric feature detection
2. Independent component representations for face recognition
3. The 'independent components' of natural scenes are edge filters
4. Blind signal separation: Statistical principles
5. Independent component analysis - a new concept?
6. An investigation into face pose distribution
7. Appearance-based object recognition using optimal feature transforms
8. Emergence of topography and complex cell properties from natural images using extensions of ica
9. Independent component analysis: Algorithms and applications
10. Information sciences
11. ICA mixture models for unsupervised classification of non-Gaussian classes and automatic context switching in blind separation
12. Comparative assessment of independent component analysis (ica) for face recognition
13. Independent component analysis of textures
14. Principal manifolds and Bayesian subspaces for visual recognition
15. Visual learning and recognition of 3-D objects from appearance
16. Subspace methods for robot vision
17. Natural image statistics and efficient coding
18. View-based and modular eigenspaces for face recognition
19. Separation of a mixture of independent sources through a maximum likelihood approach