×

Application of motion capture attributes to individual identification under corridor surveillance. (English) Zbl 1462.94057

Summary: Accurate and fast identification of a person from a security point of view is a key procedure. The most common technique of person identification uses identity cards. In contrary to the common approach we focus our research on identification based on the body movement such as the gait in this paper. The gait and the posture belong to the unique characteristics of the person that helps us to facilitate the identification. The proposed methodology allows us to incorporate personal characteristics into the access control systems using the color depth camera (RGBD). For the sake of gait analysis, the important task is to recognize the figure and extract the skeleton data from a video recording. Besides the usage of the mathematical statistics methods, we are opting to use computer animation and computer vision methods, which makes the research interdisciplinary. The main novelty of the paper is the definition and extraction of the feature vector from motion capture data, the analysis methodology and finally the selection of few statistically dominant motion attributes for the identification purposes. Besides the development of new approaches in this field, we validate proposed approaches from the perspective of accuracy.

MSC:

94A62 Authentication, digital signatures and secret sharing
94A08 Image processing (compression, reconstruction, etc.) in information and communication theory
68U10 Computing methodologies for image processing
68P01 General topics in the theory of data
68U05 Computer graphics; computational geometry (digital and algorithmic aspects)
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Acar, E., Senst, T., Kuhn, A., Keller, I., Theisel, H., Albayrak, S., and Sikora, T. 2012. Human action recognition using lagrangian descriptors. In 2012 IEEE 14th International Workshop on Multimedia Signal Processing (MMSP). 360-365.
[2] Andriluka, M., Roth, S., and Schiele, B. 2009. Pictorial structures revisited: People detection and articulated pose estimation. In 2009 IEEE Conference on Computer Vision and Pattern Recognition. 1014-1021.
[3] Barclay, C. D., Cutting, J. E., and Kozlowski, L. T. 1978. Temporal and spatial factors in gait perception that influence gender recognition. Perception & Psychophysics 23, 2, 145-152.
[4] Bazin, A. I. and Nixon, M. S. 2005. Gait verification using probabilistic methods. In Application of Computer Vision, 2005. WACV/MOTIONS ‘05 Volume 1. Seventh IEEE Workshops on. Vol. 1. 60-65.
[5] Borg, I. and Groenen, P. 2013. Modern Multidimensional Scaling: Theory and Applications. Springer Series in Statistics. Springer New York. · Zbl 0862.62052
[6] Chen, D., Wactlar, H., y. Chen, M., Gao, C., Bharucha, A., and Hauptmann, A. 2008. Recognition of aggressive human behavior using binary local motion descriptors. In 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. 5238-5241.
[7] Dubois, A. and Bresciani, J. P. 2015. Person identification from gait analysis with a depth camera at home. In 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 4999-5002.
[8] Ďurikovič, R. and Madaras, M. 2015. Controllable skeleton-sheets representation via shape diameter function. In Mathematical Progress in Expressive Image Synthesis II, H. Ochiai and K. Anjyo, Eds. Springer Japan, Tokyo, 79-90.
[9] Feng, Y., Li, Y., and Luo, J. 2016. Learning effective gait features using lstm. In 2016 23rd International Conference on Pattern Recognition (ICPR). 325-330.
[10] Gianaria, E., Grangetto, M., Lucenteforte, M., and Balossino, N. 2014. Human Classification Using Gait Features. Springer International Publishing, Cham, 16-27.
[11] Kim, J. and Kim, M. 2014. Motion capture with high-speed rgb-d cameras. In 2014 International Conference on Information and Communication Technology Convergence (ICTC). 394-395.
[12] O’haver, T. 2016. A Pragmatic Introduction to Signal Processing: With Applications in Scientific Measurement. CreateSpace Independent Publishing Platform.
[13] Peterkova, A. and Stremy, M. 2015. Obtaining the gait parameters from kinect sensor for the person identification. In 2015 IEEE 19th International Conference on Intelligent Engineering Systems (INES). 337-340.
[14] Rahman, M. W. and Gavrilova, M. L. 2017. Kinect gait skeletal joint feature-based person identification. In 2017 IEEE 16th International Conference on Cognitive Informatics Cognitive Computing (ICCI*CC). 423-430.
[15] Riečický, A., Madaras, M., Piovarci, M., and Durikovic, R. 2018. Optical-inertial synchronization of mocap suit with single camera setup for reliable position tracking. In Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP,. INSTICC, SciTePress, 40-47.
[16] Tautz, D. 1989. Hypervariability of simple sequences as a general source for polymorphic dna markers. Nucleic Acids Research 17, 16, 6463.
[17] Telford, W., Telford, W., Geldart, L., and Sheriff, R. 1990. Applied Geophysics. Monograph series. Cambridge University Press.
[18] Wang, Y., Lau, D. L., and Hassebrook, L. G. 2010. Fit-sphere unwrapping and performance analysis of 3d fingerprints. Appl. Opt. 49, 4 (Feb), 592-600.
[19] Yam, C., Nixon, M. S., and Carter, J. N. 2002. On the relationship of human walking and running: automatic person identification by gait. In Object recognition supported by user interaction for service robots. Vol. 1. 287-290 vol.1.
[20] Zhu, L., Hu, X., and Kavan, L. 2015. Adaptable anatomical models for realistic bone motion reconstruction. Comput. Graph. Forum 34, 2 (May), 459-471.
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.