1 / 41

Face recognition: Opportunities and Challenges

Microsoft Research March 17, 2006 Face recognition: Opportunities and Challenges Yu Hen Hu University of Wisconsin – Madison Dept. Electrical and Computer Enginerring Madison, WI 53706 yhhu@wisc.edu Collaborators: Nigel Boston, Charles Dyer, Weiyang Lin, Ryan Wong, Goudong Guo Agenda

niveditha
Download Presentation

Face recognition: Opportunities and Challenges

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Microsoft Research March 17, 2006 Face recognition: Opportunities and Challenges Yu Hen Hu University of Wisconsin – Madison Dept. Electrical and Computer Enginerring Madison, WI 53706 yhhu@wisc.edu Collaborators: Nigel Boston, Charles Dyer, Weiyang Lin, Ryan Wong, Goudong Guo

  2. Agenda • Human Face Recognition Problem • Recent Progress and Findings • Research at UW-Madison • Invariant of transformation group • Existing Invariants • Integral invariant • Summation invariant • Experiment Environment • Face Recognition Grand Challenge • BEE: Biometric Experimentation Environment • Preliminary FRGC Results (c) 2004-2006 by Yu Hen Hu

  3. Face Recognition Problem gallery probe image Probe (c) 2004-2006 by Yu Hen Hu

  4. Face Recognition by Human A. Adler and J. Maclean, “Performance comparison of human and automatic face recognition” Biometrics Consortium Conference 2004 (c) 2004-2006 by Yu Hen Hu

  5. Understanding Human Face Recognition • Many physiological and psychological studies have been conducted. Eg. Pawan Sinha @MIT. For example, we learned • facial configuration plays an important role in human judgments of identity [Sinha05] Celebrity faces created using identikits [Sinha05] C:\users\yuhen\1Research\face\sinha\ (c) 2004-2006 by Yu Hen Hu

  6. Human Face Recognition Limitations Harmon and Julesz, 1973 Degraded images Human can handle low resolution celebrity images quite well. [Sinha05] (c) 2004-2006 by Yu Hen Hu

  7. Face Recognition by Machine Pre-processing Feature Extraction Pattern Classification (c) 2004-2006 by Yu Hen Hu

  8. Face Recognition Process • A pattern classification problem. • Two steps: • Feature extraction • Classification • Feature extraction properties • Invariant • Discriminant • Classification • Often use distance based method Query image Feature Extraction Query image feature vector Gallery face images feature vectors Pattern Classification Recognition results Wei-Yang Lin (c) 2004-2006 by Yu Hen Hu

  9. Challenges • Sinha et al [2005] use this example to illustrate the difficulty of finding a suitable “similarity” measure to gauge similarity between a pair of faces. • In this example, the outer two faces actually belong to the same person while the middle one does not. But conventional pixel-based measures who say otherwise. • Common variations in pose (this case), lighting, expression, distance, aging remain challenges to face recognition. (c) 2004-2006 by Yu Hen Hu

  10. Face Recognition Challenges Table 1 Face Recognition Technology Evaluation Size Note that the MCINT portion of FRVT 2002 is the only test in this chart that included “video” signatures.Signatures in all other tests were a single still image. http://www.frvt.org/FRVT2002/default.htm (c) 2004-2006 by Yu Hen Hu

  11. Face Recognition Grand Challenge • Goal: to advance performance of face recognition by 10-fold (20%  2% verification rate @0.1% false alarm rate) • Focus on five different scenarios. • Status: on-going to be concluded by the end of 2005 (c) 2004-2006 by Yu Hen Hu

  12. Object Recognition • Three Approaches for object recognition (Powen Sinha) • Transformationist approach • Requires normalization • Computationally expensive. • View-based approach • Store all possible views of the same object • Expensive on storage. • Invariant-based approach • Different views  same invariant features • Desired properties of features • Invariant to variations of the same object • Discriminate to separate similar objects (c) 2004-2006 by Yu Hen Hu

  13. Euclidean Similarity Affine Projective Geometric Transformation Groups (c) 2004-2006 by Yu Hen Hu

  14. Moment Invariants • Introduced by M. K. Hu in 1962 • Advantages • Do NOT require parameterization. • Not sensitive to noise. • Limitations • Low discriminating power. • Local characteristics can NOT be extracted. (c) 2004-2006 by Yu Hen Hu

  15. Differential Invariants • Two examples in 2D • Limitation • sensitive to noise (c) 2004-2006 by Yu Hen Hu

  16. Integral Invariants • Hann and Hickman [2002] extend transformation to integrals • Advantages • do NOT require derivatives • local characteristics can be extracted • Invariants can be systematically generated • Limitation • Require analytical expression of shape (c) 2004-2006 by Yu Hen Hu

  17. Summation Invariants • Introduced by Lin et al. [2005] • Advantages • systematical approach • robustness • high discriminating power • Limitation • require parameterization (c) 2004-2006 by Yu Hen Hu

  18. Method of Moving Frame • The method of moving frame, introduce by Elie Cartan, is a tool for finding invariants under group actions. • Definition: A moving frame is a smooth, G-equivariant map (c) 2004-2006 by Yu Hen Hu

  19. Example: Differential Invariants of E(2) (c) 2004-2006 by Yu Hen Hu

  20. Example: Integral Invariants of E(2) (c) 2004-2006 by Yu Hen Hu

  21. Example: Summation Invariants of E(2) (c) 2004-2006 by Yu Hen Hu

  22. Euclidean Summation Invariants of Curves • Given a curve under Euclidean transformation • We can find a moving frame by solving (c) 2004-2006 by Yu Hen Hu

  23. Euclidean Summation Invariants of Curves • From the moving frame, a family of invariant functions can be derived (c) 2004-2006 by Yu Hen Hu

  24. Euclidean Summation Invariants of Curves • The first summation invariants are explicitly shown below where (c) 2004-2006 by Yu Hen Hu

  25. Euclidean Summation Invariants of Surfaces • Similarly, the family of surface invariants where (c) 2004-2006 by Yu Hen Hu

  26. Face Recognition Grand Challenge (FRGC) • Organized by NIST to facilitate the development of FR technology • Provide challenging problems and facial images • FRGC v2.0 dataset contains 50,000 recordings, including • high resolution still images • 3D images (c) 2004-2006 by Yu Hen Hu

  27. Four FRGC Experiments • Controlled indoor • Multiple images • 3D images • Controlled vs. uncontrolled (c) 2004-2006 by Yu Hen Hu

  28. Baseline @ FAR = 0.1% (c) 2004-2006 by Yu Hen Hu

  29. Biometric Experimentation Environment (BEE) Image Preprocessing BioBox Sub-similarity Generation Similarity Normalization Analysis (c) 2004-2006 by Yu Hen Hu

  30. Sub-similarity Generation (c) 2004-2006 by Yu Hen Hu

  31. FRGC 3D Experiment (c) 2004-2006 by Yu Hen Hu

  32. FRGC 3D Baseline Algorithm (c) 2004-2006 by Yu Hen Hu

  33. Proposed Algorithm shape curve SI PCA similarity score similarity score + PCA similarity score shape surface SI (c) 2004-2006 by Yu Hen Hu

  34. Experiment Setup • Use only 3D shape. Note 2D texture is not utilized in our experiment. • Specified the region of interest • At each pixel, compute summation invariants from a specified window region • Perform PCA to reduce dimensionality (c) 2004-2006 by Yu Hen Hu

  35. 3D Facial Surface and its Summation Invariants (c) 2004-2006 by Yu Hen Hu

  36. 3D Facial Surface and its Summation Invariants (c) 2004-2006 by Yu Hen Hu

  37. ROC Performance of Curve Invariants (c) 2004-2006 by Yu Hen Hu

  38. ROC Performance of Surface Invariants (c) 2004-2006 by Yu Hen Hu

  39. Fusion of Summation Invariants (c) 2004-2006 by Yu Hen Hu

  40. Comparison with Baseline Algorithm (c) 2004-2006 by Yu Hen Hu

  41. Conclusion • Summation invariants • novel geometric feature • provide useful shape information • fusion further improve recognition performance • In principle, one can apply summation invariants to unnormalized images (c) 2004-2006 by Yu Hen Hu

More Related