ABSTRACT
This paper addresses the problem of representing and recognizing motion trajectories. We first propose to derive scene-related equipotential lines for points in a motion trajectory and concatenate them to construct a 3D tube for representing the trajectory. Based on this 3D tube, a droplet-based method is further proposed which derives a "water droplet" from the 3D tube and recognizes trajectory activities accordingly. Our proposed 3D tube can effectively embed both motion and scene-related information of a motion trajectory while the proposed droplet- based method can suitably catch the characteristics of the 3D tube for activity recognition. Experimental results demonstrate the effectiveness of our approach.
- J. Nascimento, M. Figueiredo, and J. Marques, "Trajectory classification using switched dynamical hidden Markov models," IEEE Trans. Image Proc., pp. 1338--1348, 2010. Google ScholarDigital Library
- H. Chu, W. Lin, J. Wu, X. Zhou, Y. Chen, and H. Li, "A new heat-map-based algorithm for human group activity recognition," ACM Multimedia (MM), pp. 1069--1072, 2012. Google ScholarDigital Library
- K. Kim, D. Lee, and I. Essa, "Gaussian process regression flow for analysis of motion trajectories," ICCV, 2011.Google Scholar
- C. Rao, A. Yilmaz, and M. Shah, "View-invariant representation and recognition of actions," IJCV, 2002. Google ScholarDigital Library
- D. Ellis, E. Sommerlade, and I. Reid, "Modelling pedestrian trajectory patterns with gaussian processes," Int. Conf. Computer Vision Workshops, pp. 1229--1234, 2009.Google Scholar
- W. Lin, Y. Chen, J. Wu, H. Wang, B. Sheng, and H. Li, "A new network-based algorithm for human activity recognition in videos," IEEE Trans. CSVT, 24(5), pp. 826--841, 2014.Google Scholar
- V. Streeter, E. Wylie, and K. Bedford, "Fluid Mechanics," McGraw-Hill press, 1987.Google Scholar
- C. Chang and C. Lin. "LIBSVM: a library for support vector machines," ACM Trans. Intell. Syst. Tech., pp. 1--27, 2011. Google ScholarDigital Library
- R. Hess and A. Fern, "Discriminatively trained particle filters for complex multi-object tracking," CVPR, 2009.Google Scholar
- X. Wang, K. T. Ma, G. Ng, and E. Grimson, "Trajectory analysis and semantic region modeling using nonparametric hierarchical bayesian models, " IJCV, 2011. Google ScholarDigital Library
- B. Morris and M. Trivedi, "Trajectory learning for activity understanding: unsupervised, multilevel, and long-term adaptive approach," IEEE Trans. PAMI, 2011. Google ScholarDigital Library
Index Terms
- Representing And Recognizing Motion Trajectories: A Tube And Droplet Approach
Recommendations
Extraction and temporal segmentation of multiple motion trajectories in human motion
A new method for extraction and temporal segmentation of multiple motion trajectories in human motion is presented. The proposed method extracts motion trajectories generated by body parts without any initialization or any assumption on color ...
A Tube-and-Droplet-Based Approach for Representing and Analyzing Motion Trajectories
Trajectory analysis is essential in many applications. In this paper, we address the problem of representing motion trajectories in a highly informative way, and consequently utilize it for analyzing trajectories. Our approach first leverages the complete ...
Comments