Angry Crowds: Detecting Violent Events in Videos

Mohammadi, Sadegh; Perina, Alessandro; Kiani, Hamed; Murino, Vittorio

doi:10.1007/978-3-319-46478-7_1

Cited by 76 publications

(40 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Anomaly detection is one of the most challenging and long standing problems in computer vision [39,38,7,10,5,20,43,27,26,28,42,18,26]. For video surveillance applications, there are several attempts to detect violence or aggression [15,25,11,30] Beyond violent and non-violent patterns discrimination, authors in [38,7] proposed to use tracking to model the normal motion of people and detect deviation from that normal motion as an anomaly. Due to difficulties in obtaining reliable tracks, several approaches avoid tracking and learn global motion patterns through histogram-based methods [10], topic modeling [20], motion patterns [31], social force models [29], mixtures of dynamic textures model [27], Hidden Markov Model (HMM) on local spatio-temporal volumes [26], and context-driven method [43].…”

Section: Related Workmentioning

confidence: 99%

Real-World Anomaly Detection in Surveillance Videos

Sultani

Chen

Shah

2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

1,406

1,023

View full text Add to dashboard Cite

Surveillance videos are able to capture a variety of realistic anomalies. In this paper, we propose to learn anomalies by exploiting both normal and anomalous videos. To avoid annotating the anomalous segments or clips in training videos, which is very time consuming, we propose to learn anomaly through the deep multiple instance ranking framework by leveraging weakly labeled training videos, i.e. the training labels (anomalous or normal) are at videolevel instead of clip-level. In our approach, we consider normal and anomalous videos as bags and video segments as instances in multiple instance learning (MIL), and automatically learn a deep anomaly ranking model that predicts high anomaly scores for anomalous video segments. Furthermore, we introduce sparsity and temporal smoothness constraints in the ranking loss function to better localize anomaly during training.We also introduce a new large-scale first of its kind dataset of 128 hours of videos. It consists of 1900 long and untrimmed real-world surveillance videos, with 13 realistic anomalies such as fighting, road accident, burglary, robbery, etc. as well as normal activities. This dataset can be used for two tasks. First, general anomaly detection considering all anomalies in one group and all normal activities in another group. Second, for recognizing each of 13 anomalous activities. Our experimental results show that our MIL method for anomaly detection achieves significant improvement on anomaly detection performance as compared to the state-of-the-art approaches. We provide the results of several recent deep learning baselines on anomalous activity recognition. The low recognition performance of these baselines reveals that our dataset is very challenging and opens more opportunities for future work. The dataset is available at: https

show abstract

Section: Related Workmentioning

confidence: 99%

Real-World Anomaly Detection in Surveillance Videos

Sultani

Chen

Shah

2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

1,406

1,023

View full text Add to dashboard Cite

show abstract

“…ACC ± SD ROC-AUC STIP(HoG) + BoW [8], [9] 91.7 -STIP(HoF) + BoW [8], [9] 88.6 -MoSIFT + BoW [10] 90.9 -MoSIFT + KDE + SC [10] 94.0±1.97% 0. Method ACC ± SD ROC-AUC Jerk # [3] 95.02±0.56% -STIP (HoG) + BoW [8] 44.5% -STIP (HoF) + BoW [8] 50.5% -MoSIFT + BoW [8] 89.5% -VIF # [9] 91.31±1.06% -Interaction Force # [16] 95.51±0.79% -F L |F Cv [27] 96.89±0.21% -VIPS [25] 96 20% has been obtained. The classification result significantly benefits from the integration of the Lagrangian descriptors.…”

Section: Methodsmentioning

confidence: 99%

“…Further improvements were proposed by Xu et al [10] who substituted the bag-of-words step with a sparse coding scheme to encode MoSIFT features for violent video detection. A different approach has been proposed by Mohammadi et al with the Visual Information Processing Signature (VIPS) [25] feature. The VIPS is based on heuristic motion based rules that are related to acceleration, body compression and the aggressive drive in the video.…”

Section: Related Workmentioning

confidence: 99%

Crowd Violence Detection Using Global Motion-Compensated Lagrangian Features and Scale-Sensitive Video-Level Representation

Senst¹,

Eiselein²,

Kühn³

et al. 2017

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

Lagrangian theory provides a rich set of tools for analyzing non-local, long-term motion information in computer vision applications. Based on this theory, we present a specialized Lagrangian technique for the automated detection of violent scenes in video footage. We present a novel feature using Lagrangian direction fields that is based on a spatio-temporal model and uses appearance, background motion compensation, and long-term motion information. To ensure appropriate spatial and temporal feature scales, we apply an extended bag-of-words procedure in a late-fusion manner as classification scheme on a per-video basis. We demonstrate that the temporal scale, captured by the Lagrangian integration time parameter, is crucial for violence detection and show how it correlates to the spatial scale of characteristic events in the scene. The proposed system is validated on multiple public benchmarks and non-public, real-world data from the London Metropolitan Police. Our experiments confirm that the inclusion of Lagrangian measures is a valuable cue for automated violence detection and increases the classification performance considerably compared to stateof-the-art methods.

show abstract

“…Only few of them, though, are concerned with dangerous behaviors. These methods can be further divided into those detecting dangerous crowd behaviors, in which the individual motion is superseded by large flows as in [68,69,70,71], and those detecting closer dangerous human behaviors.…”

Section: Related Work and Datasets On Abnormal Behaviorsmentioning

confidence: 99%

Discovery and recognition of motion primitives in human activities

2019

View full text Add to dashboard Cite

We present a novel framework for the automatic discovery and recognition of motion primitives in videos of human activities. Given the 3D pose of a human in a video, human motion primitives are discovered by optimizing the ‘motion flux’, a quantity which captures the motion variation of a group of skeletal joints. A normalization of the primitives is proposed in order to make them invariant with respect to a subject anatomical variations and data sampling rate. The discovered primitives are unknown and unlabeled and are unsupervisedly collected into classes via a hierarchical non-parametric Bayes mixture model. Once classes are determined and labeled they are further analyzed for establishing models for recognizing discovered primitives. Each primitive model is defined by a set of learned parameters. Given new video data and given the estimated pose of the subject appearing on the video, the motion is segmented into primitives, which are recognized with a probability given according to the parameters of the learned models. Using our framework we build a publicly available dataset of human motion primitives, using sequences taken from well-known motion capture datasets. We expect that our framework, by providing an objective way for discovering and categorizing human motion, will be a useful tool in numerous research fields including video analysis, human inspired motion generation, learning by demonstration, intuitive human-robot interaction, and human behavior analysis.

show abstract

Angry Crowds: Detecting Violent Events in Videos

Cited by 76 publications

References 26 publications

Real-World Anomaly Detection in Surveillance Videos

Real-World Anomaly Detection in Surveillance Videos

Crowd Violence Detection Using Global Motion-Compensated Lagrangian Features and Scale-Sensitive Video-Level Representation

Discovery and recognition of motion primitives in human activities

Contact Info

Product

Resources

About