Textures of optical flow for real-time anomaly detection in crowds

Ryan, David; Denman, Simon; Fookes, Clinton; Sridharan, Sridha

doi:10.1109/avss.2011.6027327

Cited by 89 publications

(58 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…2). For detecting anomalies, we use a global GMM as in [7] and [9] to create a normality model, as it allows to model the events with a unique probability distribution. The GMM is constructed using the EM algorithm.…”

Section: B Construction Of the Gaussian Mixture Modelmentioning

confidence: 99%

“…These features are: magnitude of optical flow and uniformity (or textures of optical flow), proposed in [7] and histogram of optical flow, similarly to [4], [9], [14]. …”

Section: Feature Extractionmentioning

confidence: 99%

“…This has become the most popular approach in more recent works [2]- [4], [6]. Typical low-level features are gradients [2], [3] or features based on optical flow [4], [7].…”

Section: Introductionmentioning

confidence: 99%

“…Within this category, anomaly detection proposals are of special interest regarding the amount of works developed [l]- [4], [6], [7], [14]. In spite of this, it is still an open task due to the multiple problems that it generates.…”

Section: Introductionmentioning

confidence: 99%

“…Ryan et al [7] used a different approach. They implement a Gaussian mixture to create a normality model, using for that descriptors that incorporate flow-based information.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Detection of anomalies in surveillance scenarios using mixture models

Salgado

2017

2017 International Carnahan Conference on Security Technology (ICCST)

View full text Add to dashboard Cite

Abstract-In this paper we present a robust and simple method for the detection of anomalies in surveillance scenarios. We use a "bottom-up" approach that avoids any object tracking, making the system suitable for anomaly detection in crowds. A robust optical flow method is used for the extraction of accurate spatio-temporal motion information, which allows to get simple but discriminative descriptors that are employed to train a Gaussian mixture model. We evaluate our system in a publicly available dataset, concluding that our method outperforms similar anomaly detection approaches but with a simpler model and low-sized descriptors.

show abstract

Section: B Construction Of the Gaussian Mixture Modelmentioning

confidence: 99%

“…These features are: magnitude of optical flow and uniformity (or textures of optical flow), proposed in [7] and histogram of optical flow, similarly to [4], [9], [14]. …”

Section: Feature Extractionmentioning

confidence: 99%

“…This has become the most popular approach in more recent works [2]- [4], [6]. Typical low-level features are gradients [2], [3] or features based on optical flow [4], [7].…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…Ryan et al [7] used a different approach. They implement a Gaussian mixture to create a normality model, using for that descriptors that incorporate flow-based information.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations