Computation time is a fundamental concern when tracking objects in real time, especially in complex scenes. Inspired by previous works on automatic failure detection and in situ evaluation of tracking, the authors propose in this study an adaptive tracking algorithm based on pattern recognition techniques, which uses more computing resources only when tracking is likely to fail. Tracking quality is discretised into two binary values and a supervised classifier is trained using some features obtained from the tracking itself and ground truth data. During the operation of the classifier, whenever the tracking quality diminishes, the tracking algorithm reacts in a predefined way in order to avoid the failure. Two specific examples are presented, in which the action taken is different when a potential risk situation is detected: either the number of particles increases or the algorithm used to track changes. The experimental work shows that these methods can be easily implemented with a substantial reduction of processing time but with little tracking performance loss.

References

1. 1)
  - 5. Kembhavi, A., Schwartz, W.R., Davis, L.S.: ‘Resource allocation for tracking multiple targets using particle filters’. Eighth Int. Workshop on Visual Surveillance, 2008.
2. 2)
  - 1. Zhou, S., Chellappa, R., Moghaddam, B.: ‘Visual tracking and recognition using appearance-adaptive models in particle filters’, IEEE Trans. Image Process., 2004, 13, (11), pp. 1491–1506 (doi: 10.1109/TIP.2004.836152).
3. 3)
  - 7. Morris, B., Trivedi, M.: ‘Improved vehicle classification in long traffic video by cooperating tracker and classifier modules’. IEEE Int. Conf. Advanced Video and Signal Based Surveillance (AVSS), 2006.
4. 4)
  - 18. Medrano, C., Martínez, J., Igual, R., Herrero, J., Orrite, C.: ‘Gaussian approximation for tracking occluding and interacting targets’, J. Math. Imaging Vis., 2010, 36, (3), pp. 241–253 (doi: 10.1007/s10851-009-0183-9).
5. 5)
  - 4. Straka, O., Simandl, M.: ‘Particle filter with adaptive sample size’, Kybernetika, 2011, 47, (3), pp. 385–400.
6. 6)
  - 3. Soto, A.: ‘Self adaptive particle filter’. Int. Joint Conf. Artificial Intelligence (IJCAI), 2005, pp. 1398–1406.
7. 7)
  - 2. Fox, D.: ‘KLD-sampling: adaptive particle filters and mobile robot localization’. Advances in Neural Information Processing Systems (NIPS), 2001.
8. 8)
  - 11. Powers, R.M., Pao, L.Y.: ‘Detecting track-loss for the probabilistic data association filter in the absence of truth data’. 43rd IEEE Conf. Decision and Control, 2004, 4, pp. 4316–4323.
9. 9)
  - 14. Medrano, C., Herrero, J.E., Martinez, J., Orrite, C.: ‘Mean field approach for tracking similar objects’, Comput. Vis. Image Underst., 2009, 113, (8), pp. 907–920 (doi: 10.1016/j.cviu.2009.03.007).
10. 10)
  - 13. Liu, R., Li, S.Z., Yuan, X., He, R.: ‘Online determination of track loss using template inverse matching’. Eighth Int. Workshop on Visual Surveillance – VS2008, 2008.
11. 11)
  - 6. Rehrl, T., Theibing, M., Bannat, A., et al: ‘Tracking using Bayesian inference with a two-layer graphical model’. Proc. Int. Conf. Image Processing (ICIP 2010), 2010, pp. 3961–3964.
12. 12)
  - 9. Lei, Y., Ding, X., Wang, S.: ‘Visual tracker using sequential Bayesian learning: discriminative, generative and hybrid’, IEEE Trans. Syst. Man Cybern., B, 2008, 38, (6), pp. 1578–1591 (doi: 10.1109/TSMCB.2008.928226).
13. 13)
  - 10. Pao, L.Y., Khawsuk, W.: ‘Determining track loss without truth information for distributed target tracking applications’, Proc. American Control Conf., 2000, vol. 6, pp. 4363–4367.
14. 14)
  - 16. Zhao, T., Nevatia, R., Wu, b.: ‘Segmentation and tracking of multiple humans in crowded environments’, IEEE Trans. Pattern Anal. Mach. Intell., 2008, 30, (7), pp. 1198–1211 (doi: 10.1109/TPAMI.2007.70770).
15. 15)
  - 12. Wu, H., Sankaranarayanan, A., Chellappa, R.: ‘In situ evaluation of tracking algorithms using time reversed chain’. IEEE Conf. Computer Vision and Pattern Recognition, 2007.
16. 16)
  - 8. Powell, G., Marshall, D., Smets, P., Ristic, B., Maskell, S.: ‘Joint tracking and classification of airbourne objects using particle filters and the continuous transferable belief model’. Ninth Int. Conf. Information Fusion, 2006.
17. 17)
  - 15. Medrano, C., Igual, R., Orrite, C., Plaza, I.: ‘Occlusion management in sequential mean field Monte Carlo methods, pattern recognition and image analysis’, Lect. Notes Comput. Sci., 2011, 6669, pp. 444–451 (doi: 10.1007/978-3-642-21257-4_55).
18. 18)
  - 17. Khan, Z., Balch, T., Dellaert, F.: ‘MCMC-based particle filtering for tracking a variable number of interacting targets’, IEEE Trans. Pattern Anal. Mach. Intell., 2005, 27, (11), pp. 1805–1819 (doi: 10.1109/TPAMI.2005.223).

Adaptive tracking algorithms to improve the use of computing resources

References

Related content