Timely, robust crowd event characterization

Kaltsa, Vagia; Briassouli, Alexia; Kompatsiaris, Ioannis; Strintzis, Michael G.

doi:10.1109/icip.2012.6467455

Cited by 12 publications

(10 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our approach, the delay in the detection of some frames after the event occurs is because of our strategy of detection, in which an abnormal event is detected if the temporal stability is becoming below the dynamic threshold (defined as half the average of temporal stabilities of previous frames). This requires some times to be detected, which Table 4 Performance of our proposed crowd change detection method in terms of recall and precision using UMN dataset compared to [11,19,26] Approach in [19] 84.75 100 justifies the delay. At the same time, this strategy is suitable to avoid false alarms.…”

Section: Crowd Change Detectionmentioning

confidence: 95%

“…To demonstrate the effectiveness of our proposed approach, we compare our results to adjacency-matrix based clustering (AMC) method [11], spatial temporal co-occurrence Gaussian mixture models (STCOG) method [26], and to the method proposed in [19], which is based on dense optical flow and particle advection. The precision and recall of all these methods are listed in Table 4.…”

Section: Crowd Change Detectionmentioning

confidence: 96%

“…Given all these difficulties, conventional human detection or tracking paradigms fail in such cases. To overcome this problem, alternative solutions which consist of tracking particles [19,20,29] or local features [3,11,18,23] instead of pedestrians have been proposed. Other methods operate on foreground masks and consider them as the regions of interest [5,6,9], called activity area in [6].…”

Section: Crowd Trackingmentioning

confidence: 99%

“…This problem covers different subproblems such as crowd change or anomaly detection [5,6,11,18,20,26], and crowd event recognition or characterization [2,8,10,17,19,23,29], in which the goal is to automatically detect changes or to alternatively recognize crowd events in video sequences. In general, there are three main categories of crowd behavior analysis methods.…”

Section: Related Workmentioning

confidence: 99%

“…For quantitive evaluation, we employ the relative mean frame error metric proposed in [19]. It is defined as:…”

Section: Crowd Change Detectionmentioning

confidence: 99%

See 4 more Smart Citations

Spatial and temporal variations of feature tracks for crowd behavior analysis

Fradi

Dugelay

2015

J Multimodal User Interfaces

View full text Add to dashboard Cite

The study of crowd behavior in public areas or during some public events is receiving a lot of attention in security community to detect potential risk and to prevent overcrowd. In this paper, we propose a novel approach for change detection, event recognition and characterization in human crowds. It consists of modeling time-varying dynamics of the crowd using local features. It also involves a feature tracking step which allows excluding feature points on the background and extracting long-term trajectories. This process is favourable for the later crowd event detection and recognition since the influence of features irrelevant to the underlying crowd is removed and the tracked features undergo an implicit temporal filtering. These feature tracks are further employed to extract regular motion patterns such as speed and flow direction. In addition, they are used as an observation of a probabilistic crowd function to generate fully automatic crowd density maps. Finally, the variation of these attributes (local density, speed, and flow direction) in time is employed to determine the ongoing crowd behaviors. The experimental results on two different datasets demonstrate the effectiveness of our proposed approach for early detection of crowd change and accurate results for event recognition and characterization.

show abstract

Section: Crowd Change Detectionmentioning

confidence: 95%

Section: Crowd Change Detectionmentioning

confidence: 96%

Section: Crowd Trackingmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

“…For quantitive evaluation, we employ the relative mean frame error metric proposed in [19]. It is defined as:…”

Section: Crowd Change Detectionmentioning

confidence: 99%

See 3 more Smart Citations

Spatial and temporal variations of feature tracks for crowd behavior analysis

Fradi

Dugelay

2015

J Multimodal User Interfaces

View full text Add to dashboard Cite

show abstract

Machine Learning Methods for Behaviour Analysis and Anomaly Detection in Video

Isupova¹

2018

Springer Theses

View full text Add to dashboard Cite

Behaviour analysis and anomaly detection are key components of intelligent vision systems. Anomaly detection can be considered from two perspectives: abnormal events can be defined as those that violate typical activities or as a sudden change in behaviour. Topic modeling and change point detection methodologies, respectively, are employed to achieve these objectives.The thesis starts with development of novel learning algorithms for a dynamic topic model. Topics extracted by the learning algorithms represent typical activities happening within an observed scene. These typical activities are used for likelihood computation. The likelihood serves as a normality measure in anomaly detection decision making. A novel anomaly localisation procedure is proposed. In the considered dynamic topic model a number of topics, i.e., typical activities, should be specified in advance. A novel dynamic nonparametric hierarchical Dirichlet process topic model is then developed where the number of topics is determined from data. Conventional posterior inference algorithms require processing of the whole data through several passes. It is computationally intractable for massive or sequential data. Therefore, batch and online inference algorithms for the proposed model are developed. A novel normality measure is derived for decision making in anomaly detection. The latter part of the thesis considers behaviour analysis and anomaly detection within the change point detection methodology. A novel general framework for change point detection is introduced. Gaussian process time series data is considered and a change is defined as an alteration in hyperparameters of the Gaussian process prior. The problem is formulated in the context of statistical hypothesis testing and several tests suitable both for offline and online data processing and multiple change point detection are proposed. Theoretical properties of the proposed tests are derived based on the distribution of the test statistics. i ACKNOWLEDGMENTS

show abstract

Background

Isupova¹

2018

Machine Learning Methods for Behaviour Analysis and Anomaly Detection in Video

View full text Add to dashboard Cite

Timely, robust crowd event characterization

Cited by 12 publications

References 13 publications

Spatial and temporal variations of feature tracks for crowd behavior analysis

Spatial and temporal variations of feature tracks for crowd behavior analysis

Machine Learning Methods for Behaviour Analysis and Anomaly Detection in Video

Background

Contact Info

Product

Resources

About