Density aware anomaly detection in crowded scenes

Download

index.pdf

Date

2016-08-01

Author

GÜNDÜZ, Ayse Elvan
Öngün, Cihan
Temizel, Tugba Taskaya
Temizel, Alptekin

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

43
views

0
downloads

Coherent nature of crowd movement allows representing the crowd motion using sparse features. However, surveillance videos recorded at different periods of time are likely to have different crowd densities and motion characteristics. These varying scene properties necessitate use of different models for an effective representation of behaviour at different periods. In this study, a density aware approach is proposed to detect motion-based anomalies for scenes having varying crowd densities. In the training, the sparse features are modelled using separate hidden Markov models, each of which becomes an expert for specific scene characteristics. These models are then used for anomaly detection. The proposed method automatically adapts to the changing scene dynamics by switching to the most representative model at each frame. The authors demonstrate the effectiveness and real-time performance of the proposed method on real-life datasets as well as on simulated crowd videos that they generated and made publicly available to download.

Subject Keywords

Event detection

URI

https://hdl.handle.net/11511/32142

Journal

IET COMPUTER VISION

DOI

https://doi.org/10.1049/iet-cvi.2015.0345

Collections

Graduate School of Informatics, Article

Suggestions

OpenMETU
Core

Event Boundary Detection Using Audio Visual Features and Web casting Texts with Imprecise Time Information MÜJDAT, Bayar; ALAN, Özgür; SAMET, Akpınar; ORKUNT, Sabuncu; Çiçekli, Fehime Nihan; Alpaslan, Ferda Nur (2010-07-21) We propose a method to detect events and event boundaries in soccer videos by using web-casting texts and audio-visual features. The events and their inaccurate time information given in web-casting texts need to be aligned with the visual content of the video. We overcome this issue by utilizing textual, visual and audio features. Existing methods assume that the time at which the event occurs is given precisely (in seconds). However, most web-casting texts presented by popular organizations such as uefa.c...
Abnormal Crowd Behavior Detection Using Novel Optical Flow-Based Features Direkoglu, Cem; Sah, Melike; O'Connor, Noel E. (2017-09-01) In this paper, we propose a novel optical flow based features for abnormal crowd behaviour detection. The proposed feature is mainly based on the angle difference computed between the optical flow vectors in the current frame and in the previous frame at each pixel location. The angle difference information is also combined with the optical flow magnitude to produce new, effective and direction invariant event features. A one-class SVM is utilized to learn normal crowd behavior. If a test sample deviates si...
Gray Level Topological Angle Detection of High Curvature Points Cihan, Ibrahim Kivanc; ŞENEL, HAKAN GÜRAY (2010-07-07) A new method of angle detection based on the topological median filter is proposed. Topological opening operator is used to detect corners and Topological closing operator is used to calculate the corner angles on gray level images.
Gray level topological corner detection Cihan, Ibrahim Kivanc; Senel, Hakan Gueray (2007-06-07) A new corner detection algorithm based on the topological median filter is proposed. Topological quasi dilation and erosion operators with circular structuring element are used to detect the corners on gray level images.
Hand-crafted versus learned representations for audio event detection Kucukbay, Selver Ezgi; Yazıcı, Adnan; Kalkan, Sinan (2022-04-01) Audio Event Detection (AED) pertains to identifying the types of events in audio signals. AED is essential for applications requiring decisions based on audio signals, which can be critical, for example, for health, surveillance and security applications. Despite the proven benefits of deep learning in obtaining the best representation for solving a problem, AED studies still generally employ hand-crafted representations even when deep learning is used for solving the AED task. Intrigued by this, we investi...

Citation Formats

A. E. GÜNDÜZ, C. Öngün, T. T. Temizel, and A. Temizel, “Density aware anomaly detection in crowded scenes,” IET COMPUTER VISION, pp. 374–381, 2016, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/32142.