Evaluation of Matching Metrics for Trajectory-Based Indexing and Retrieval of Video Clips

Khalid, Shehzad; Naftel, Andrew

doi:10.1109/acvmot.2005.49

Cited by 18 publications

(18 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous work has sought to represent moving object trajectories through a wide variety of direction schemes, polynomial models and other function approximations [1][2][3][4][5][6][7][8][9][10][19][20][21][22]. The importance of selecting the most appropriate trajectory model has received relatively scant attention [11]. It is also surprising to find that many of these candidate indexing schemes have not yet been applied to the problem of motion data mining and trajectory classification.…”

Section: Background and Related Workmentioning

confidence: 99%

“…To achieve dimensionality reduction, we consider object trajectories as motion time series and use a coefficient indexing scheme. A performance comparison of different motion indexing schemes can be found in [11].…”

Section: Learning Trajectory Patterns Using Self-organizing Mapsmentioning

confidence: 99%

“…Much of the earlier research focus has been on high-level object trajectory representation schemes that are able to produce compressed forms of motion data [1][2][3][4][5][6][7][8][9][10][11]. This work presupposes the existence of some low-level tracking scheme for reliably extracting object-based trajectories [12,13].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Classifying spatiotemporal object trajectories using unsupervised learning of basis function coefficients

Khalid

Naftel

2005

Proceedings of the Third ACM International Workshop on Video Surveillance &Amp; Sensor Networks

View full text Add to dashboard Cite

This paper proposes a novel technique for clustering and classification of object trajectory-based video motion clips using spatiotemporal functional approximations. A Mahalanobis classifier is then used for the detection of anomalous trajectories. Motion trajectories are considered as time series and modeled using the leading Fourier coefficients obtained by a Discrete Fourier Transform. Trajectory clustering is then carried out in the Fourier coefficient feature space to discover patterns of similar object motions. The coefficients of the basis functions are used as input feature vectors to a Self-Organising Map which can learn similarities between object trajectories in an unsupervised manner. Encoding trajectories in this way leads to efficiency gains over existing approaches that use discrete point-based flow vectors to represent the whole trajectory. Experiments are performed on two different datasets -synthetic and pedestrian object tracking -to demonstrate the effectiveness of our approach. Applications to motion data mining in video surveillance databases are envisaged.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Section: Learning Trajectory Patterns Using Self-organizing Mapsmentioning

confidence: 99%

See 1 more Smart Citation

Classifying spatiotemporal object trajectories using unsupervised learning of basis function coefficients

Khalid

Naftel

2005

Proceedings of the Third ACM International Workshop on Video Surveillance &Amp; Sensor Networks

View full text Add to dashboard Cite

show abstract

“…2. The distance d rms in (6) represents a quantity related to the area between P Q and P S , normalized with respect to the length of the interval of integration, and can be evaluated as a closed-form expression, e.g., in [26], for m = 3. The distance functions d max and d min can be evaluated through a line search technique.…”

Section: B Modified K-means Algorithm For Trajectory Clusteringmentioning

confidence: 99%

Detection and Classification of Highway Lanes Using Vehicle Motion Trajectories

Melo

Naftel

Bernardino

et al. 2006

IEEE Trans. Intell. Transport. Syst.

Self Cite

118

View full text Add to dashboard Cite

Abstract-Intelligent vision-based traffic surveillance systems are assuming an increasingly important role in highway monitoring and road management schemes. This paper describes a low-level object tracking system that produces accurate vehicle motion trajectories that can be further analyzed to detect lane centers and classify lane types. Accompanying techniques for indexing and retrieval of anomalous trajectories are also derived. The predictive trajectory merge-and-split algorithm is used to detect partial or complete occlusions during object motion and incorporates a Kalman filter that is used to perform vehicle tracking. The resulting motion trajectories are modeled using variable low-degree polynomials. A K-means clustering technique on the coefficient space can be used to obtain approximate lane centers. Estimation bias due to vehicle lane changes can be removed using robust estimation techniques based on Random Sample Consensus (RANSAC). Through the use of nonmetric distance functions and a simple directional indicator, highway lanes can be classified into one of the following categories: entry, exit, primary, or secondary. Experimental results are presented to show the real-time application of this approach to multiple views obtained by an uncalibrated pan-tilt-zoom traffic camera monitoring the junction of two busy intersecting highways.

show abstract

“…Much of the earlier research focus in motion analysis has been on high-level object trajectory representation schemes that are able to produce compressed forms of motion data (Aghbari et al, 2003;Chang et al, 1998;Dagtas et al, 2000;Hsu & Teng, 2002;Jin & Mokhtarian, 2004;Khalid & Naftel, 2005;Shim & Chang, 2004). This work presupposes the existence of some low-level visual tracking scheme for reliably extracting object-based trajectories (Hu, Tan, Wang & Maybank, 2004;Vlachos et al, 2002).…”

Section: Introductionmentioning

confidence: 99%

Behavior Recognition Using any Feature Space Representation of Motion Trajectories

Khalid¹

2012

Recent Developments in Video Surveillance

View full text Add to dashboard Cite

Evaluation of Matching Metrics for Trajectory-Based Indexing and Retrieval of Video Clips

Cited by 18 publications

References 17 publications

Classifying spatiotemporal object trajectories using unsupervised learning of basis function coefficients

Classifying spatiotemporal object trajectories using unsupervised learning of basis function coefficients

Detection and Classification of Highway Lanes Using Vehicle Motion Trajectories

Behavior Recognition Using any Feature Space Representation of Motion Trajectories

Contact Info

Product

Resources

About