access icon free Low-resolution face alignment and recognition using mixed-resolution classifiers

A very common case for law enforcement is recognition of suspects from a long distance or in a crowd. This is an important application for low-resolution face recognition (in the authors' case, face region below 40 × 40 pixels in size). Normally, high-resolution images of the suspects are used as references, which will lead to a resolution mismatch of the target and reference images since the target images are usually taken at a long distance and are of low resolution. Most existing methods that are designed to match high-resolution images cannot handle low-resolution probes well. In this study, they propose a novel method especially designed to compare low-resolution images with high-resolution ones, which is based on the log-likelihood ratio (LLR). In addition, they demonstrate the difference in recognition performance between real low-resolution images and images down-sampled from high-resolution ones. Misalignment is one of the most important issues in low-resolution face recognition. Two approaches – matching-score-based registration and extended training of images with various alignments – are introduced to handle the alignment problem. Their experiments on real low-resolution face databases show that their methods outperform the state-of-the-art.

Inspec keywords: image classification; statistical analysis; image registration; face recognition; police data processing

Other keywords: LLR; image recognition; law enforcement; log-likelihood ratio; suspect recognition; matching-score-based registration; face recognition; low-resolution face alignment; suspect high-resolution images; mixed-resolution classifiers; low-resolution face databases; image extended training

Subjects: Image recognition; Computer vision and image processing techniques; Other topics in statistics; Public administration; Other topics in statistics

References

    1. 1)
      • 16. Wang, X., Tang, X.: ‘Hallucinating face by eigen transformation’, IEEE Trans. Syst. Man Cybern., Appl. Rev., 2005, 35, (3), pp. 425434.
    2. 2)
      • 35. Grgic, M., Delac, K., Grgic, S.: ‘SCface – surveillance cameras face database’, Multimedia Tools Appl., 2011, 51, pp. 863879.
    3. 3)
      • 32. Spreeuwers, L.J., Boom, B.J., Veldhuis, R.N.J.: ‘Better than best: matching score based face registration’. Proc. of the 28th Symp. on Information Theory in the Benelux, Enschede, May 2007, pp. 125132.
    4. 4)
      • 3. Biswas, S., Bowyer, K., Flynn, P.: ‘Multidimensional scaling for matching low-resolution face images’, IEEE Trans. Pattern Anal. Mach. Intell., 2011, 34, (10), pp. 20192030.
    5. 5)
      • 23. Turk, M., Pentland, A.: ‘Face recognition using eigen faces’. Proc. 1991 IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, June 1991, pp. 586591.
    6. 6)
      • 1. Zhang, D., He, J., Du, M.: ‘Morphable model space based face super-resolution reconstruction and recognition’, Image Vis. Comput., 2012, 30, pp. 100108.
    7. 7)
      • 2. Zou, W., Yuen, P.: ‘Very low resolution face recognition problem’. 2010 Fourth IEEE Int. Conf. on Biometrics: Theory Applications and Systems (BTAS), 2010, pp. 16.
    8. 8)
      • 25. Ahonen, T., Hadid, A., Pietikainen, M.: ‘Face description with local binary patterns: application to face recognition’, IEEE Trans. Pattern Anal. Mach. Intell., 2006, 28, (12), pp. 20372041.
    9. 9)
      • 4. Lei, Z., Liao, S., Jain, A., et al: ‘Coupled discriminant analysis for heterogeneous face recognition’, IEEE Trans. Inf. Forensics Sec., 2012, 7, (6), pp. 17071716.
    10. 10)
      • 30. Min, J., Bowyer, K., Flynn, P.: ‘Eye perturbation approach for robust recognition of inaccurately aligned faces’. Audio- and Video-Based Biometric Person Authentication, 2005 (LNCS, 3546), pp. 4150.
    11. 11)
      • 22. Phillips, P., Flynn, P., Scruggs, T., et al: ‘Overview of the face recognition grand challenge’. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, 2005 CVPR 2005, June 2005, vol. 1, pp. 947954.
    12. 12)
      • 5. Ren, C., Dai, D., Yan, H.: ‘Coupled kernel embedding for low-resolution face image recognition’, IEEE Trans. Image Process., 2012, 21, (8), pp. 37703783.
    13. 13)
      • 12. Huang, H., He, H.: ‘Super-resolution method for face recognition using nonlinear mappings on coherent features’, IEEE Trans. Neural Netw., 2011, 22, (1), pp. 121130.
    14. 14)
      • 11. Pong, K.-H., Lam, K.-M.: ‘Multi-resolution feature fusion for face recognition’, Pattern Recognit., 2014, 47, (2), pp. 556567.
    15. 15)
      • 27. Chen, D., Cao, X., Wang, L., et al: ‘Bayesian face revisited: a joint formulation’. Computer Vision – ECCV 2012, 2012 (LNCS, 7574), pp. 566579.
    16. 16)
      • 28. Bazen, A., Veldhuis, R.: ‘Likelihood-ratio-based biometric verification’, IEEE Trans. Circuits Syst. Video Technol., 2004, 14, (1), pp. 8694.
    17. 17)
      • 17. Moutafis, P., Kakadiaris, I.A.: ‘Semi-coupled basis and distance metric learning for cross-domain matching: application to low-resolution face recognition’. IEEE Int. Joint Conf. on Biometrics, Clearwater, FL, 2014, pp. 18.
    18. 18)
      • 21. Xu, X., Liu, W., Li, L.: ‘Face hallucination: how much it can improve face recognition’. 2013 3rd Australian Control Conf. (AUCC), November 2013, pp. 9398.
    19. 19)
      • 18. Baker, S., Kanade, T.: ‘Hallucinating faces’. . Proc. Fourth IEEE Int. Conf. on Automatic Face and Gesture Recognition, 2000, 2000, pp. 8388.
    20. 20)
      • 19. Hennings-Yeomans, P., Baker, S., Kumar, B.: ‘Recognition of low-resolution faces using multiple still images and multiple cameras’. Second IEEE Int. Conf. on Biometrics: Theory, Applications and Systems, 2008 BTAS 2008, 2008, pp. 16.
    21. 21)
      • 7. Peng, Y., Spreeuwers, L.J., Veldhuis, R.N.J.: ‘Likelihood ratio based mixed resolution facial comparison’. Third Int. Workshop on Biometrics and Forensics (IWBF2015), Gjøvik, Norway, March 2015, pp. 15.
    22. 22)
      • 15. Gunturk, B., Batur, A., Altunbasak, Y., et al: ‘Eigenface-domain super-resolution for face recognition’, IEEE Trans. Image Process., 2003, 12, (5), pp. 597606.
    23. 23)
      • 6. Zou, W., Yuen, P.: ‘Very low resolution face recognition problem’, IEEE Trans. Image Process., 2012, 21, (1), pp. 327340.
    24. 24)
      • 33. Rowley, H.A., Baluja, S., Kanade, T.: ‘Human face detection in visual scenes’. Advances in Neural Information Processing Systems, 1996, vol. 8, pp. 875881.
    25. 25)
      • 14. Jiang, J., Hu, R., Han, Z., et al: ‘Graph discriminant analysis on multi-manifold (GDAMM): a novel super resolution method for face recognition’. 2012 19th IEEE Int. Conf. on Image Processing (ICIP), September 2012, pp. 14651468.
    26. 26)
      • 10. Hennings-Yeomans, P., Kumar, B., Baker, S.: ‘Robust low-resolution face identification and verification using high resolution features’. 2009 16th IEEE Int. Conf. on Image Processing (ICIP), 2009, pp. 3336.
    27. 27)
      • 29. Vera-Rodriguez, R., Tome, P., Fierrez, J., et al: ‘Comparative analysis of the variability of facial landmarks for forensics using CCTV images’. Proc. Image and Video Technology: Sixth Pacific-Rim Symp., PSIVT 2013, October 28–November 1 2013, Guanajuato, Mexico, 2014, pp. 409418.
    28. 28)
      • 31. Boom, B.J., Spreeuwers, L.J., Veldhuis, R.N.J.: ‘Automatic face alignment by maximizing similarity score’. Proc. of the Seventh Int. Workshop on Pattern Recognition in Information Systems, Biosignals, Madeira, Portugal, June 2007, pp. 221230.
    29. 29)
      • 9. Zhou, C., Zhang, Z., Yi, D., et al: ‘Low-resolution face recognition via simultaneous discriminant analysis’. 2011 Int. Joint Conf. on Biometrics (IJCB), October 2011, pp. 16.
    30. 30)
      • 34. Cognitec Systems GmbH. FaceVACS-SDK Version 8.7.0.2, 2015.
    31. 31)
      • 26. Belhumeur, P., Hespanha, J., Kriegman, D.: ‘Eigenfaces vs. Fisherfaces: recognition using class specific linear projection’, IEEE Trans. Pattern Anal. Mach. Intell., 1997, 19, (7), pp. 711720.
    32. 32)
      • 24. Veldhuis, R.N.J., Bazen, A.M.: ‘One-to-template and one-to-one verification in the single- and multi-user case’. 26th Symp. on Information Theory in the Benelux, Brussels, Belgium, Brussels, May 2005, pp. 3946.
    33. 33)
      • 20. Hu, S., Maschal, R., Young, S.S., et al: ‘Face recognition performance with super resolution’, Appl. Opt., 2012, 51, (18), pp. 42504259.
    34. 34)
      • 36. Viola, P., Jones, M.: ‘Rapid object detection using a boosted cascade of simple features’. Proc. of the 2001 IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, 2001 CVPR 2001, 2001, vol. 1, pp. I–511I–518.
    35. 35)
      • 13. Li, B., Chang, H., Shan, S., et al: ‘Low-resolution face recognition via coupled locality preserving mappings’, IEEE Signal Process. Lett., 2010, 17, (1), pp. 2023.
    36. 36)
      • 8. Bilgazyev, E., Efraty, B., Shah, S., et al: ‘Improved face recognition using super-resolution’. 2011 Int. Joint Conf. on Biometrics (IJCB), 2011 Int. Joint Conf. on, October 2011, pp. 17.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-bmt.2016.0026
Loading

Related content

content/journals/10.1049/iet-bmt.2016.0026
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading