During video surveillance, face re-identification allows recognition and targeting of individuals of interest from faces captured across a network of video cameras. In such applications, face recognition is challenging because faces are captured under limited spatial and temporal constraints. In addition, facial models for recognition are commonly designed using a limited number of representative reference samples from faces captured under specific conditions, regrouped into facial trajectories. Given new reference samples (provided by an operator or through some self-updating process), updating facial models may allow maintenance of a high level of performance over time. Although adaptive ensembles have been successfully applied to robust modelling of an individual's facial appearance, reference data samples from a trajectory must be stored for validation. In this study, a memory management strategy based on Kullback–Leiber (KL) divergence is proposed to rank and select the most relevant validation samples over time in adaptive individual-specific ensembles. When new reference samples become available for an individual, updates to the corresponding ensemble are validated using a mixture of new and previously-stored samples. Only the samples with the highest KL divergence are preserved in memory for future adaptations. This strategy is compared with reference classifiers using videos from the face in action data. Simulation results show that the proposed strategy tends to select discriminative samples from wolf-like individuals for validation. It allows maintenance of a high level of performance, while reducing the number of samples per individual by up to 80%.

References

1. 1)
  - 10. Viola, P., Jones, M.: ‘Robust real-time face detection’, Int. J. Comput. Vis., 2004, 2, (57), pp. 137–154 (doi: 10.1023/B:VISI.0000013087.49260.fb).
2. 2)
  - 16. De-la Torre, M., Granger, E., Sabourin, R., Gorodnichy, D.O.: ‘An individual-specific strategy for management of reference data in adaptive ensembles for person re-identification’. Int. Conf. on Imaging for Crime Detection and Prevention, London, UK, December 2013, pp. 1–7.
3. 3)
  - 7. Goh, R., Liu, L., Liu, X., Chen, T.: ‘The CMU face in action database’. Analysis and Modeling of Faces and Gestures, Beiging, China, October 2005, pp. 255–263.
4. 4)
  - 22. Ojala, T., Pietikainen, M., Maenpaa, T.: ‘Multiresolution gray-scale and rotation invariant texture classification with local binary patterns’, Trans. Pattern Anal. Mach. Intell., 2002, 24, (7), pp. 971–987 (doi: 10.1109/TPAMI.2002.1017623).
5. 5)
  - 6. Freni, B., Marcialis, G., Roli, F.: ‘Template selection by editing algorithms: a case study in face recognition’. Int. Association of Pattern Recognition, Orlando, USA, December 2008, vol. 5342, pp. 745–754.
6. 6)
  - 14. Ekenel, H.K., Szasz-Toth, L., Stiefelhagen, R.: ‘Open-set face recognition-based visitor interface system’. Computer Vision Systems, Liege, Belgium, October 2009, vol. 5815, pp. 43–52.
7. 7)
  - 17. Guo, X., Yin, Y., Dong, C., Yang, G., Zhou, G.: ‘On the class imbalance problem’. Int. Conf. on Computing, Networking and Communications, Piscataway, USA, August 2008, vol. 4, pp. 192–201.
8. 8)
  - 4. Kubat, M., Matwin, S.: ‘Addressing the curse of imbalanced training sets: one-sided selection’. Int. Conf. Machine Learning, Nashville, USA, July 1997, pp. 179–186.
9. 9)
  - 15. Pagano, C., Granger, E., Sabourin, R., Gorodnichy, D.O.: ‘Detector ensembles for face recognition in video surveillance’. Int. Joint Conf. on Neural Networks, Brisbane, Australia, June 2012, pp. 1–8.
10. 10)
  - 19. Kachites McCallum, A., Nigam, K.: ‘Employing EM and pool-based active learning for text classification’. Int. Conf. Machine Learning, San Francisco, USA, July 1998, pp. 350–358.
11. 11)
  - 9. Li, F., Wechsler, H.: ‘Open set face recognition using transduction’, Trans. Pattern Anal. Mach. Intell., 2005, 27, (11), pp. 1686–1697 (doi: 10.1109/TPAMI.2005.224).
12. 12)
  - 2. De-la Torre, M., Granger, E., Radtke, P.V.W., Sabourin, R., Gorodnichy, D.O.: ‘Incremental update of biometric models in face-based video surveillance’. Int. Joint Conf. Neural Networks, Brisbane, Australia, June 2012, pp. 1–8.
13. 13)
  - 13. Kamgar-Parsi, B., Lawson, W., Kamgar-Parsi, B.: ‘Toward development of a face recognition system for watchlist surveillance’, Trans. Pattern Anal. Mach. Intell., 2011, 33, (10), pp. 1925–1937 (doi: 10.1109/TPAMI.2011.68).
14. 14)
  - 5. Polikar, R., Udpa, L., Udpa, S.S., Honavar, V.: ‘Learn++: an incremental learning algorithm for MLP networks’, Int. Conf. Syst. Man Cybern., 2001, 31, (4), pp. 497–508 (doi: 10.1109/5326.983933).
15. 15)
  - 3. De la Torre, M., Granger, E., Radtke, P.V.W., Sabourin, R., Gorodnichy, D.O.: ‘Self-updating with facial trajectories for video-to-video face recognition’. Int. Conf. on Pattern Recognition, Stockholm, Sweden, 24–28 August 2014, pp. 1–8.
16. 16)
  - 8. Doddington, G., Liggett, W., Martin, A., Przybocki, M., Reynolds, D.: ‘Sheep, goats, lambs and wolves: a statistical analysis of speaker performance’. Int. Conf. Spoken Language Processing, Sidney, Australia, December 1998, pp. 1351–1354.
17. 17)
  - 21. Lim, C.P., Harrison, R.F.: ‘An incremental adaptive network for on-line supervised learning and probability estimation’. Neural Networks, Houston, USA, June 1997, pp. 925–939.
18. 18)
  - 1. Connolly, J.F., Granger, E., Sabourin, R.: ‘Evolution of heterogeneous ensembles through dynamic particle swarm optimization for video-based face recognition’, Pattern Recogn., 2012, 45, pp. 2460–2477 (doi: 10.1016/j.patcog.2011.12.016).
19. 19)
  - 18. Scheffer, T., Decomain, C., Wrobel, S.: ‘Active hidden Markov models for information extraction’. Int. Symp. on Intelligent Data Analysis, Berlin, Germany, June 2001, vol. 2189, pp. 309–318.
20. 20)
  - 12. Jain, A.K., Ross, A.: ‘Learning user-specific parameters in a multibiometric system’. Int. Conf. Image Processing, Rochester, USA, September 2002, pp. 57–60.
21. 21)
  - 11. Matta, F., Dugelay, J.-L.: ‘Person recognition using facial video information: a state of the art’, J. Vis. Lang. Comput., 2009, 20, (3), pp. 180–187 (doi: 10.1016/j.jvlc.2009.01.002).
22. 22)
  - 20. Dagan, I., Engelson, S.P.: ‘Committee-based sampling for training probabilistic classifiers’. Int. Conf. Machine Learning, San Francisco, USA, July 1995, pp. 150–157.

Individual-specific management of reference data in adaptive ensembles for face re-identification

References

Related content