Abstract
In this paper, we present an improved methodology for detecting and tracking the various posture and movement of people in a crowded and dynamic environment with the help of a single RGB-D camera. The RGB-D cameras are also called as low depth cameras or ranging cameras. The depth camera provides depth information for each pixel. The indigenous RGB-D pixels are transformed into a new point ensemble image (PEI) and human detection and tracking in a 3D space can be accomplished in a more effective and accurate manner. PEI representation, unlike height map representation, projects all the points in the cell into the grid. First, the detector locates the human physically from the probable candidates who are then carefully filtered in a supervised learning and classification second stage. The statistics of color and height are then computed for associating data to generate the 3D orientation of the tracked individuals. We use classifiers such as JHCH and HOHD. The statistics of color and height are then computed for associating data to generate the 3D orientation of the tracked individuals. In tracking, we try to estimate the similarity criteria in order to compare the current frame and the detected response. We have used RANSAC matching algorithm in the tracking stage. The qualitative and quantitative experiments are performed using the different datasets that show a promising improvement by improving the accuracy of the system to 97%. We have concentrated on the false positives and miss rates in the detecting stage and track lost error and ID switch error in the tracking stage. We have produced significant improvements by reducing these errors and even works well in a highly occluded environment. As we concentrate only on the upper part of the body, occlusion is not going to affect our system anywhere. The results of both qualitative and quantitative experiments obtained show a promising improvement by reducing track lost error thereby improving accuracy.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Jun Liu, Ye Liu, Guyue Zhang, Peiru Zhu, Yan Qiu Chen, “Detection and tracking people in real time with RGB-D camera”, Elsevier, 2014, pp. 16–23.
I. Haritaoglu, M. Flickner, D.Beymer, “Video-crm: understanding customers behaviors in stores, in: SPIE Proceeding”, Video surveillance and transportation imaging application, 2013.
Zhenghao Xi, Dongmei Xu, Wanqing Song, Yang Zheng, “A* algorithm with dynamic weights for multiple objects tracking”, Elsevier, 2015.
Qiang Zhu, Shai Avidian, Mei-Chen Yeh, and Kwang-Ting Cheng, “Fast human detection using a cascade of histogram of oriented gradients”, IEEE Conf. on Computer vision and pattern recognition, 2006.
K.Kim, V.Lincent Lepetit, W.Woo, “Real time interactive modeling and scalable multiple object tracking for AR”, Elsevier, 2012, pp. 945–954.
A.H.Maziran, “A powerful approach to real-time mobile objects tracking in crowded environments utilizing 3Dvideo analysis of fixed camera, particle filter and natural network”, Elsevier, 2013, pp. 6357–6365.
Harish Bhaskar, Kartik Durivedi, Debi Prosad Drogra, Mohammed Al-Mualla, Lyudmila Mihaylova, “Autonomous detection and tracking under illumination changes occlusion and moving camera”, Elsevier, 2015, pp. 343–354.
Z.He,Y.Cui, Hongpeng Wang, X.You, P.Chen, “One global optimization method in network flow model for multiple object tracking”, Elsevier, 2015, pp. 21–32.
Bogdan Kwalek, Michal Kepski, “Human fall detection on embedded platform using depth maps and wireless accelerometer”, Elsevier, 2014, pp. 489–501.
Olga Zoidi, Nikos Nikolaidis, Anastasio Tefas, Loannis Pitas, “Stereo object tracking with fusion of texture, color and disparity information”, Elsevier, 2014, pp. 573–589.
A.H.Maziran, “A powerful approach to real-time mobile objects tracking in crowded environments utilizing 3Dvideo analysis of fixed camera, particle filter and natural network”, Elsevier, 2013, pp. 6357–6365.
Houari Sabirin, Munchurl Kim, “Low complexity object detection and tracking with inter-layer graph mapping and intra-layer graph refinement in H.264/SVC bit streams”, Elsevier, 2013, pp. 1531–1539.
W.Choi, C.Pantofaru, Silvio Savarese, “A general framework for tracking multiple people from a moving camera”, IEEE Trans. Pattern Anal. Mach. Intell, 2013, pp. 1577–1591.
Kiyoung Kim, Vincent Lepetit, Woontack Woo,“Real time interactive modeling and scalable multiple object tracking for AR”, Elsevier, 2012, pp. 945–954.
B.K. Dan, Y.S. Kim, Suryanto,J.Y. Jung, S.J. Ko, “Robust people counting system based on sensor fusion”, IEEE Trans. Consum. Electron, 2012, pp. 1013–1021.
L.Xia, C.C.Chen, J.K Aggarwal, “Human detection using depth information by kinect”, IEEE Conference on computer vision and pattern recognition workshop, 2011, pp. 15–22.
K.Levi, Y. Weiss, “Learning object detection from a small number of examples: the importance of good features”, IEEE Conf. Computer vision and pattern recognition, 2013, pp. 53–60.
A.Ess,B.Leibe, K.Schindler, LV.Gool, “Robust multiperson tracking from a mobile platform”, IEEE Trans. Pattern Anal. Mach. Intell, 2009, pp. 1831–1846.
M.Enzweiler, A.Eigenstetter, B.Scheile, D.M.Gavrila, “Multi cue pedestrian classification with partial occlusion handling”, IEEE Conf. on computer vision and pattern recognition, 2010, pp. 990–997.
M.Luber, L.Spinello, K.O.Arras, “People tracking in RGB-D data with on-line boosted target models”, IEEE Conf. Intelligent robots and system, 2011.
N.Dalal, B.Triggs, “Histograms of oriented gradients for human detection”, IEEE Conf. on computer vision and patterns recognition, 2005.
F.Fleuret, J.Berclaz, R.Lengage, P.Fua, “Multicamera people tracking with a probabilistic occupancy map”, IEEE Trans.Pattern Anal.Mach.Intell, 2008, pp. 267–282.
Q.Zhu, M.C.Yeh, K.T.Cheng, S.Avidan, “Fast human detection using a cascade of histograms of Oriented gradients”, IEEE Conference on Computer Vision and Pattern Recognition, 2006, pp. 1491–1498.
P.Felzenszwalb, “Learning models for object recognition”, IEEE Conf. on computer vision and pattern recognition, 2001, pp. 1056–1062.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Mahalakshmi, M., Kanthavel, R., Hemavathy, N. (2019). Real-Time Human Detection and Tracking Using PEI Representation in a Dynamic 3D Environment. In: Bhaskar, M., Dash, S., Das, S., Panigrahi, B. (eds) International Conference on Intelligent Computing and Applications. Advances in Intelligent Systems and Computing, vol 846. Springer, Singapore. https://doi.org/10.1007/978-981-13-2182-5_19
Download citation
DOI: https://doi.org/10.1007/978-981-13-2182-5_19
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-13-2181-8
Online ISBN: 978-981-13-2182-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)