Abstract
This paper focuses on human behavior recognition where the main problem is to bridge the semantic gap between the analogue observations of the real world and the symbolic world of human interpretation. For that, a fusion architecture based on the Transferable Belief Model framework is proposed and applied to action recognition of an athlete in video sequences of athletics meeting with moving camera. Relevant features are extracted from videos, based on both the camera motion analysis and the tracking of particular points on the athlete’s silhouette. Some models of interpretation are used to link the numerical features to the symbols to be recognized, which are running, jumping and falling actions. A Temporal Belief Filter is then used to improve the robustness of action recognition. The proposed approach demonstrates good performance when tested on real videos of athletics sports videos (high jumps, pole vaults, triple jumps and long jumps) acquired by a moving camera and different view angles. The proposed system is also compared to Bayesian Networks.
Similar content being viewed by others
Notes
Many other papers are available on Smets’ homepage http://iridia.ulb.ac.be/~psmets. The web page also proposes links towards other researchers in the TBM community as well as softwares.
The software can bedownloaded onhttp://www.irisa.fr/Vista/Motion2D.
In the first version, 18 points aretracked; but in this paper, image quality is not sufficient for such a level ofdetail (which is not useful here).
The notion of distinctness is close, but notequivalent to the independence notion in probability theory. See [43] for more details.
Here, the frame is an image butnot a FoD.
De Morgan’s algebra can also beapplied.
A focal element corresponds to a propositionfor which the belief is not null.
A focal set in a BBA is a set for which the associated beliefmass is not null.
When the BBAs are defined on a FoD witha cardinality greater than 2, the decision must not be taken using beliefmasses, but pignistic probabilities [52].
References
Hu W, Tan T, Wang L, Maybank S (2004) A survey on visual surveillance of object motion and behaviors. IEEE Trans Systems Man Cybern C 34(3):334–352
Messer K, Christmas WJ, Jaser E, Kittler J, Levienaise-Obadia B, Koubaroulis D (2005) A unified approach to the generation of semantic cues for sports video annotation. Signal Processing 85:357–383
Jaimes A, Sebe N (2005) Multimodal human computer interaction: a survey. In: IEEE International Workshop on Human Computer Interaction in conjunction with ICCV, vol 3766. Beijing, China, pp 1–15
Li B, Errico JH, Pan H, Sezan I (2004) Bridging the semantic gap in sports video retrieval and summarization. J Vis Commun Image Represent 15:393–424
Sadlier DA, O’Connor NE (2005) Event detection in field sports video using audio-visual features and a support vector machine. IEEE Trans Circuit Syst Video Technol 15(10):1225–1233
Lew M, Sebe N, Eakins J (2002) Challenges in image and video retrieval. Lect Notes Comput Sci ICIVR 2383:1–6
Freitas ND, Brochu E, Barnard K, Duygulu P, Forsyth D (2002) Bayesian models for massive multimedia databases: a new frontier. In: Valencia International Meeting on Bayesian Statistics/2002 ISBA International Meeting
Shah M (2003) Understanding human behavior from motion imagery. Mach Vis Appl 14:210–214
Hongeng S, Nevatia R, Bremond F (2004) Video-based event recognition and probabilistic recognition methods. Comput Vis Image Underst 96:129–162
Klir GJ, Wierman MJ (1999) Uncertainty-based information. Elements of generalized information theory, 2nd edn. Studies in fuzzyness and soft computing. Physica-Verlag, New York
Dubois D, Grabisch M, Prade H, Smets Ph (2001) Using the transferable belief model and a qualitative possibility theory approach on an illustrative example: the assessment of the value of a candidate. Int J Intell Syst 16:1245–1272
Ristic B, Smets P (2005) Target identification using belief functions and implication rules. IEEE Trans Aerosp Electron Syst 41(3):1097–1102
Petkovic M, Jonker W (2004) Integrated use of different content derivation techniques within a multimedia database management system. J Vis Commun Image Represent 15:303–329
Smets P, Kennes R (1994) The transferable belief model. Artif Intell 66(2):191–234
Shafer G (1976) A mathematical theory of evidence. Princeton University Press, Princeton
Smets P (1994) Advances in the Dempster–Shafer theory of evidence—What is Dempster–Shafer’s model? In: Yager RR, Fedrizzi M, Kacprzyk J, edition. Wiley, New York, pp 5–34
Ramasso E, Rombaut M, Pellerin D (2006) A temporal belief filter improving human action recognition in videos. In: IEEE international conference on acoustics, speech and signal processing, vol 2, pp 141–144
Schubert J (2004) Clustering belief functions based on attracting and conflicting metalevel evidence using Potts spin mean field theory. Inform Fusion 5(4):309–318
Wang L, Hu W, Tan T (2003) Recent developments in human motion analysis. Pattern Recognit 36(3):585–601
Yacoob Y, Black M (1999) Parametrized modeling and recognition of activities. Comput Vis Image Underst 73(2):232–247
Rabiner LR (1989) A tutorial on hidden Markov models and selected applications in speech recognition. Proc IEEE 77:257–285
Murphy KP (2002) Dynamic Bayesian Networks: representation, inference and learning. PhD thesis, UC Berkeley (CSD)
Xiang T, Gong S (2003) Discovering Bayesian causality among visual events in a complex outdoor scene. In: IEEE on advanced video and signal based surveillance, pp 177–182
Luo Y, Wu TD, Hwang JN (2003) Object-based analysis and interpretation of human motion in sports video sequences by dynamic Bayesian networks. Comput Vis Image Underst 92:196–216
Vefghi L, Linkens DA (1999) Dynamic monitoring and control of patient anaesthetic and dose levels: time-delay, moving-average neural networks, and principal components analysis. Comput Methods Programs Biomed 59:91–106
Zhong D, Chang S-F (2004) Real-time view recognition and event detection for sports video. J Vis Commun Image Represent 15:330–347
Medioni G, Cohen I, Brémond F, Hongeng S, Nevatia R (2001) Event detection and analysis from video streams. PAMI 23(8):873–889
Ayers D, Shah M (2001) Monitoring human behavior from video taken in an office environment. Image Vis Comput 13:833–846
Ding Z, Bunke H, Schneider M, Kandel A (2005) Fuzzy timed Petri net: definitions, properties, and applications. Math Comput Model 41:345–360
Lee SJ, Seong PH (2004) Development of automated operating procedure system using fuzzy colored Petri nets for nuclear power plants. Ann Nucl Energy 31:849–869
Fay A (2000) A fuzzy knowledge-based system for railway traffic control. Eng Appl Artif Intell 13:719–729
Bourbakis N, Gattiker JR, Bebis G (2003) Interpreting a dynamic and uncertain world: task-based control. Int J Artif Intell Tools 12(1):5–85
Nigro JM, Loriette-Rougegrez S, Rombaut M (2002) Driving situation recognition with uncertainty management and rule-based systems. Eng Appl Artif Intell 15:217–228
Rombaut M, Jarkass I, Denoeux T (1999) State recognition in discret dynamical systems using Petri nets and evidence theory. In: European conference on symbolic and quantitative approaches to reasoning with uncertainty
Girondel V, Caplier A, Bonnaud L, Rombaut M (2005) Belief theory-based classifiers comparison for static human body postures recognition in video. Int J Signal Process 2(1):29–33
Hammal Z, Caplier A, Rombaut M (2005) Belief theory applied to facial expressions classification. In: International conference on advances in pattern recognition, Bath, UK
Moeslund TB, Granum E (2001) A survey of computer vision-based human motion capture. Comput Vis Image Underst 81:231–268
Wang J, Singh S (2003) Video analysis of human dynamics—a survey. Real-Time Imaging 9(5):321–346
Odobez JM, Bouthemy P (1995) Robust multiresolution estimation of parametric motion models. J Vis Commun Image Represent 6(4):348–365
Piriou G, Bouthemy P, Peyrard N, Yao JF (2002) Probabilistic models of image motion for recognition of dynamic content in video. In: International workshop on computer vision and image analysis, vol 11. Las Palmas de Gran Canaria, Spain
Fablet R, Bouthemy P, Perez P (2002) Non parametric motion characterization using causal probabilistic models for video indexing and retrieval. IEEE Trans Image Process 11(4):393–407
Panagiotakis C, Tziritas G (2004) Recognition and tracking of the members of a moving human body. In: Articulated motion and deformable objects, pp 86–98
Yaghlane B, Smets P, Mellouli K (2002) Independence concept for belief functions. In: Technologies for constructing intelligent systems: tools, Heidelberg, Germany. Physica-Verlag GmbH, New York
Ayoun A, Smets Ph (2001) Data association in multi-target detection using the transferable belief model. Int J Intell Syst 16(10):1167–1182
Zribi M, Benjelloun M (2003) Parametric estimation of Dempster–Shafer belief functions. In: International conference on information fusion, pp 485–491
Elouedi Z, Mellouli K, Smets Ph (2004) Assessing sensor reliability for multisensor data fusion within the transferable belief model. IEEE Trans Syst Man Cybern 34(1):782–787
Smets P (1993) Beliefs functions: the disjunctive rule of combination and the Generalized Bayesian Theorem. Int J Approx Reason 9:1–35
Rombaut M, Zhu YM (2002) Study of Dempster–Shafer theory for image segmentation applications. Image Vis Comput 20(1):15–23
Denoeux T, Yaghlane AB (2002) Approximating the combination of belief functions using the fast moebius transform in a coarsened frame. Int J Approx Reason 37:77–101
Philippe Smets (2002) The application of the matrix calculus to belief functions. Int J Approx Reason 31(1–2):1–30
Haenni R, Lehmann N (2003) Implementing belief function computations. Int J Intell Syst 18(1):31–49
Smets Ph (2005) Decision making in the TBM: the necessity of the pignistic transformation. Int J Approx Reason 38:133–147
Ramasso E, Pellerin D, Rombaut M (2006) Belief scheduling for the recognition of human action sequence. In: International conference on information fusion, Florence, Italia, pp 1–8
Panagiotakis C, Ramasso E, Tziritas G, Rombaut M, Pellerin D (2006) Shape-motion based athlete tracking for multilevel action recognition. In: Perales FJ, Fisher RB (eds) Proceedings of the fourth international conference on articulated motion and deformable objects. Springer, Berlin, pp 385–394
Witten IH, Frank E (2005) Data-mining: practical machine learning tools and techniques, 2nd edn. Morgan Kaufman, San Francisco, Publishers, pp 560. ISBN 0-12-088407-0
Acknowledgments
This research is partially supported by SIMILAR European excellence network. The authors thank the Vista research team at Irisa/Inria Rennes (France) for the use of the Motion2D software.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ramasso, E., Panagiotakis, C., Pellerin, D. et al. Human action recognition in videos based on the Transferable Belief Model. Pattern Anal Applic 11, 1–19 (2008). https://doi.org/10.1007/s10044-007-0073-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10044-007-0073-y