Recognition of Human Behavior Based on Chaos Invariant and Relevance Vector Machine

doi:10.3969/j.issn.1007-2985.2014.03.009

Abstract

Abstract: An algorithm is put forward for the recognition of human behavior based on chaos invariant characteristics and relevance vector machine(RVM).First,the motion track generated by the joints of human is extracted to represent the nonlinear system of human movement behavior and C-C method is used to estimate the delay to obtain the dimension of phase space reconstituted by every motion track.Furthermore,the chaos invariant which represents human behavior is extracted from the phase space and RVM algorithm is used to recognize human behavior.Finally KTH,Weizmann human behavior database and ballet database are applied to test the effect of the algorithm,and the result proves that this method has the better recognition effect than others.

Key words: chaos system, behavior recognition, chaos invariant, RVM

DAI Zhi-Qiang, DONG Jian-Feng. Recognition of Human Behavior Based on Chaos Invariant and Relevance Vector Machine[J]. journal6, 2014, 35(3): 37-43.

References

[1] JHUANG.H,SERRE T,WOLF L,et al.Biologically Inspired System for Action Recognition[C].Proceedings of IEEE 11th International Conference.Taichung,Taiwan:ICCV,2007:342-346.

[2] SCHINDLER K,VAN G L.Action Snippets:How Many Frames Does Action Recognition Require?[C].Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Anchorage,Alaska,USA:CVPR,2008:1 121-1 127.

[3] IKIZLER N,FORSYTH D.Searching Video for Complex Activities with Finite State Models[C].Proceedings of IEEE Conf. Computer Vision and Pattern Recognition.Minneapolis,MN,USA:CVPR,2007:489-496.

[4] LAXTON B,LIM J,KRIEEGMAN L D.Temporal Contextual and Ordering Constraints for Recognizing Complex Activities in Video[C].Proceedings of IEEE Conf. Computer Vision and Pattern Recognition.Minneapolis,MN USA:CVPR,2007:298-306.

[5] LAPTV I,LINDEBER T.Space-Time Interest Points[C].Proceedings of Ninth IEEE International Conf. on Computer Vision.Nice,France:ICCV,2003:432-439.

[6] LIU Jingen,MUBARAK SHAH.Learning Human Actions via Information Maximization[C].Proceedings of IEEE Conf. on Computer Vision and Pattern Recognition.Anchorage,AK,USA:CVPR,2008:812-819.

[7] NIEBLES J C,LI Feifei.A Hierarchical Model of Shape and Appearance for Human Action Classification[C].Proceedings of IEEE Conf. on Computer Vision and Pattern Recognition Minneapolis.MN USA:CVPR,2007:65-73.

[8] ALE S,BASHARAT A,SHAH M.Chaotic Invariants for Human Action Recognition[C].Proceedings of IEEE Computer Vision and Pattern Recognition.Rio de Janeiro,Brazil:ICCV,2007:290-298.

[9] GONG Wenjuan,ANDREW D,BAGDANOV F,et al.Automatic Key Pose Selection for 3D Human Action Recognition[J].Computer Science,2010,61(6):290-299.

[10] REN Haibing,XU Guangyou.Articulated-Model Based Upper-Limb Pose Estimation[C].Proceedings of IEEE International Symposium on Computational Intelligence in Robotics and Automation.Aanff,Alberta,Canada:CIRA,2001:450-454.

[11] XU Feng,KIN M L,DAI Qinghai.Video-Object Segmentation and 3D-Trajectory Estimation for Monocular Video Sequences[J].Image and Vision Computing,2011,29(1):190-205.

[12] PACKAR N H,CRUTCHFIELD J P,FARMER J D,et al.Geometry from a Time Series[J].Phys. Rev. Lett.,1980,45(9):712-716.

[13] KIM H S,EYKHOLT R,SALAS J D.Nonlinear Dynamics,Delay Times,and Embedding Windows[J].Physica D:Nonlinear Phenomena,1999,127(1):48-60.

[14] GRASSBERGER P.Grassberger-Procaccia Algorithm[J].Scholarpedia,2007,2(5):3 043-3 044.

[15] LI Xibing,WANG Qisheng,YAO Jinrui,et al.Time Series Prediction for Surrounding Rock’s Deformation of Mine Lanes in Soft Rock[J].Journal of Central South University of Technology,2008,15(2):224-229.

[16] ROSENSTEIN M T,COLLINS J J,LUCA D C J.A Practical Method for Calculating Largest Lyapunov Exponents from Small Data Sets[J].Physica D,1993,65（1）:117-134.

[17] TIPPING M E.Sparse Bayesian Learning and the Relevance Vector Machine[J].Journal of Machine Learning Research,2001,1(6):211-244.

[18] FATHI A,GREG MORI.Action Recognition by Learning Mid-Level Motion Features[C].Proceedings of IEEE CS Conf. Computer Vision and Pattern Recognition.Anchorage,AK,USA:CVPR,2008:311-317.