Journal of KIISE:Software and Applications (한국정보과학회논문지:소프트웨어및응용)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

Illumination Robust Face Recognition using Ridge Regressive Bilinear Models

Ridge Regressive Bilinear Model을 이용한 조명 변화에 강인한 얼굴 인식

  • 신동수 (LG전자 전자기술원) ;
  • 김대진 (포항공과대학교 컴퓨터공학과) ;
  • 방승양 (포항공과대학교 컴퓨터공학과)
  • Published : 2007.01.15
Download PDF
⟨ Previous Next ⟩

Abstract

The performance of face recognition is greatly affected by the illumination effect because intra-person variation under different lighting conditions can be much bigger than the inter-person variation. In this paper, we propose an illumination robust face recognition by separating identity factor and illumination factor using the symmetric bilinear models. The translation procedure in the bilinear model requires a repetitive computation of matrix inverse operation to reach the identity and illumination factors. Sometimes, this computation may result in a nonconvergent case when the observation has an noisy information. To alleviate this situation, we suggest a ridge regressive bilinear model that combines the ridge regression into the bilinear model. This combination provides some advantages: it makes the bilinear model more stable by shrinking the range of identity and illumination factors appropriately, and it improves the recognition performance by reducing the insignificant factors effectively. Experiment results show that the ridge regressive bilinear model outperforms significantly other existing methods such as the eigenface, quotient image, and the bilinear model in terms of the recognition rate under a variety of illuminations.

얼굴 인식 시스템의 성능은 조명 변화로 인하여 발생하는 개인내 (intra-person) 차이가 개인간 (inter-person)의 차이보다 클 수 있기 때문에 조명 변화에 많은 영향을 받는다. 본 연구에서는 이러한 문제를 해결하기 위해서 대칭형 bilinear 모델을 이용하여 조명 요소와 신원 요소를 분리하는 방법을 제안한다. Bilinear 모델로 조명 요소와 신원 요소를 얻기 위한 translation 과정은 반복적 역행렬을 구하는 것이 요구되는데 입력 데이타에 따라 수렴하지 않는 경우가 발생할 수 있다. 이러한 문제를 완화하기 위해서 ridge regression 모델과 bilinear 모델을 결합한 ridge regressive bilinear 모델을 제안하였다. 제안된 모델은 조명 요소와 신원 요소의 분산을 적절히 줄여줌으로서 bilinear 모델에 안정성을 제공하며, 인식에 더 많은 고차원 요소 정보를 이용하게 함으로써 인식 성능을 높여 준다. 실험 결과에서 제안한 ridge regressive bilinear 모델이 bilinear 모델, 고유얼굴(eigenface) 방법, Quotient image 보다 좋은 인식 성능을 보여줌을 확인 할 수 있다.

Keywords

References

  1. Tammy Riklin-Raviv and Amnon Shashua, 'The quotient image: class-based rerendering and recognition with varying illuminations,' IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 23, issue 2, pp. 129-139, 2001 https://doi.org/10.1109/34.908964
  2. Shaohua Zhou and Rama Chellappa, 'Rank constrained recognition under unknown illuminations,' Proceedings of Analysis and Modeling of Faces and Gestures, pp. 11-18, 2003
  3. Jiali Zhao, Yong Su, Dejun Wang and Siwei Luo, 'Illumination ratio image: ynthesizing and recognition with varying illuminations,' Pattern Recognition Letters, vol. 23, issue 15, pp. 2703-2710, 2003
  4. Dang-Hui Lui, Kin-Man Lam and Lan-Sun Shen, 'Illumination invariant face recognition,' Pattern Recognition, vol. 38, pp. 1705-1716, 2005 https://doi.org/10.1016/j.patcog.2005.03.009
  5. Shiguang Shan, Wen Gao, Bo Cao and Debin Zhao, 'Illumination normalization for robust face recognition against varying lighting conditions,' Proceedings of Analysis and Modeling of Faces and Gestures, pp. 157, 2003
  6. Peter N. Belhumeur and David J. Kriegman, 'What Is the Set of Images of an object under all possible illumination conditions?,' International Journal of Computer Vision, vol. 28, num. 3, pp. 245-260, 1998 https://doi.org/10.1023/A:1008005721484
  7. Athinodoros S. Georghiades, Kriegman. D. J and Peter N. Belhumeur, 'Illumination cones for recognition under variable lighting: Faces,' Computer Vision and Pattern Recognition, p.52, 1998
  8. Athinodoros S. Georghiades, Kriegman. D. J and Peter N. Belhumeur, 'From few to many: Illumination cone models for face recognition under variable lighting and pose,' IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 23, issue 6, pp. 643-660, 2001 https://doi.org/10.1109/34.927464
  9. Ronen Basri and David Jacobs, 'Lambertian reflectance and linear subspace,' IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 25, issue 2, pp. 218-233, 2003 https://doi.org/10.1109/TPAMI.2003.1177153
  10. Ronen Basri and David Jacobs, 'Photometric stereo with general, unknown lighting,' Computer Vision and Pattern Recognition, vol. 2, pp. 374, 2001
  11. Haitao Wang, Stan Li, Yangsheng Wang and Weiwei Zhang, 'Illumination modeling and normalization for face recognition', Proceedings of Analysis and Modeling of Faces and Gestures, pp. 104-111, 2003
  12. Joshua B. Tenenbaum, 'Separating style and content with bilinear models,' Neural Computation, vol. 12, pp. 1247-1283, 2000 https://doi.org/10.1162/089976600300015349
  13. Tormod Naes and Bjorn-Helge Mevik, 'Understanding the collinearity problem in regression and discriminant analysis,' Journal of Chemometrics, vol. 15, issue 4, pp. 413-426, 2001 https://doi.org/10.1002/cem.676
  14. Trevor Hastie, Robert Tibshirani and Jerome Friedman, 'The Elements of Statistical Learning Data Mining, Inference and Prediction,' Springer, pp. 59-64, 2001
  15. Maring Hoy, Frank Westad and Harald Martens, 'Combining bilinear modelling and ridge regression', Journal of Chemometrics, vol. 16, issue 6, pp. 313-318, 2002 https://doi.org/10.1002/cem.727
  16. T. Sim, S. Baker and M. Bsat, 'The CMU pose, illumination, and expression (PIE) database,' IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 25, issue 12, pp. 1615-1618, 2003 https://doi.org/10.1109/TPAMI.2003.1251154