Trajectory Analysis for Sport and Video Surveillance

Meneses, Yuri López de; Roduit, Pierre; Luisier, Florian; Jacot, Jacques

doi:10.1142/9789812834461_0020

Cited by 3 publications

(2 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, a trajectory represents the geometric and temporal information that can be of key importance for some analyses. Trajectories can be analyzed in their spatial, temporal or spatiotemporal dimension by projecting them into the corresponding subspaces [16]. Analysis of the object trajectories offers the possibility of extracting information on the behavior of the moving objects and their interactions with the environment.…”

Section: A Tracking Of Moving Objects and Analysis Of Their Trajectomentioning

confidence: 99%

3D Dynamic Scene Surveillance and Management Using a 3D Kinetic Spatial Data Structure

Mostafavi

Beni

Gavrilova

2009

2009 International Conference on Advanced Geographic Information Systems &Amp; Web Services

View full text Add to dashboard Cite

Tracking, monitoring and management of moving objects become increasingly important in modern geospatial application. An important example of such applications is monitoring and management of a dynamic scene captured by a network of videos or sensors for ensuring for example the security of the people during a public event, or for leading emergency or rescue teams in real environment. While current GIS are widely used to represent and manage spatiotemporal phenomena, their applications for 3D dynamic environments are very limited due to the 2D static nature of most of existing spatial data structures. In this paper, we first present a review of important requirements for efficient management of a dynamic scene. Next, we present a 3D kinetic data structure based on Delaunay tetrahedralization and Voronoi diagram that allows representation of static and moving objects and their interactions in a 3D dynamic environment, simultaneous tracking of a large number of moving objects, detection of the important events and collisions and other interesting analysis of a dynamic scene. Finally, we discuss other potentials and limitations of the proposed model and suggest some new research avenues regarding the improvement of the proposed model.

show abstract

Section: A Tracking Of Moving Objects and Analysis Of Their Trajectomentioning

confidence: 99%

3D Dynamic Scene Surveillance and Management Using a 3D Kinetic Spatial Data Structure

Mostafavi

Beni

Gavrilova

2009

2009 International Conference on Advanced Geographic Information Systems &Amp; Web Services

View full text Add to dashboard Cite

show abstract

“…After autocalibrating a camera and metric rectifying the input trajectories, path models are constructed from these trajectories and similarity measure is adopted to compare the input trajectory with the path model for abnormal activity detection. Ariel [6] proposed an approach using spatial information for associating trajectories from multiple views, which are represented as consecutive points of a joint ground plane in the world coordinate system. Methods based on pure geometric constraints heavily rely on the accuracy during the correspondence process.…”

Section: Introductionmentioning

confidence: 99%

Global Trajectory Construction across Multi-cameras via Graph Matching

Zhu

Liu

Wang

et al. 2011

2011 Sixth International Conference on Image and Graphics

View full text Add to dashboard Cite

Behavior analysis across multi-cameras becomes more and more popular with the rapid development of camera network in video surveillance. In this paper, we propose a novel unsupervised graph matching framework to associate trajectories across partially overlapping cameras. Firstly, trajectory extraction is based on object extraction and tracking and is followed by a homographic projection to a mosaic-plane. And we extract appearance and spatio-temporal features for trajectory description. Then a robust graph matching algorithm based on reweighted random walk is adopted for trajectory association. The association is formulated as node ranking and selection on an association graph whose nodes represent candidate correspondences of trajectories. Finally, the pairs of corresponding trajectories in overlapping regions are fused by an adaptive averaging scheme, in which trajectories with more observations and longer length is given higher weight. Experiments and comparison on real scenarios demonstrate the effectiveness of the proposed approach.

show abstract