A Crowd Analysis Framework for Detecting Violence Scenes

Gkountakos, Konstantinos; Ioannidis, Konstantinos; Tsikrika, Theodora; Vrochidis, Stefanos; Kompatsiaris, Ioannis

doi:10.1145/3372278.3390725

Cited by 21 publications

(12 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are two methods namely: optical flow [15] [16] and spatial-time gradient [17] [18]. Optical flow presents motion properly; however, it is driven by huge overheads.…”

Section: Methodology Of Our Approachmentioning

confidence: 99%

Beyond Tracking in Crowd: Analyzing Crowd based on Physical Characteristics

Siraj¹

2020

Preprint

View full text Add to dashboard Cite

The safety of people is an important phenomenon nowadays. This importance arises due to the crowded places including subway station, universities, colleges, airport, shopping mall and square, and city squares. Therefore, the development of an effective system based on physical characteristics of crowd layout is of significant demand. In this paper, we proposed a novel automated and intelligent systems for crowd event analysis based on a set of physical elements. For this purpose, we take into account optical flow and spatial-time gradient, contour features, and Gaussian processes. Our method combine these characteristics into a unique model to deal with the challenging problem of crowd event analysis. For evaluating our proposed method, we consider a benchmark dataset and a number of different performance metrics. These analysis demonstrate the robustness and effectiveness of our proposed method.

show abstract

“…There are two methods namely: optical flow [15] [16] and spatial-time gradient [17] [18]. Optical flow presents motion properly; however, it is driven by huge overheads.…”

Section: Methodology Of Our Approachmentioning

confidence: 99%

Beyond Tracking in Crowd: Analyzing Crowd based on Physical Characteristics

Siraj¹

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…(i) object detection, which focuses on identifying and locating a predefined set of objects of interest (see, e.g., Bochkovskiy et al, 2020), (ii) face recognition, which is able to identify specific individuals on the basis of their faces (see, e.g., Learned-Miller et al, 2016), (iii) activity recognition, which involves recognizing actions of interest performed for instance by humans and vehicles (see, e.g., Jobanputra et al, 2019), and (iv) crowd violence detection, which focuses on detecting outbreaks of crowd violence (see, e.g., Gkountakos et al, 2020).…”

Section: Visual Analysis Processesmentioning

confidence: 99%

Real‐time threat assessment based on hidden Markov models

et al. 2023

Self Cite

View full text Add to dashboard Cite

An essential factor toward ensuring the security of individuals and critical infrastructures is the timely detection of potentially threatening situations. To this end, especially in the law enforcement context, the availability of effective and efficient threat assessment mechanisms for identifying and eventually preventing crime‐ and terrorism‐related threatening situations is of utmost importance. Toward this direction, this work proposes a hidden Markov model‐based threat assessment framework for effectively and efficiently assessing threats in specific situations, such as public events. Specifically, a probabilistic approach is adopted to estimate the threat level of a situation at each point in time. The proposed approach also permits the reflection of the dynamic evolution of a threat over time by considering that the estimation of the threat level at a given time is affected by past observations. This estimation of the dynamic evolution of the threat is very useful, since it can support the decisions by security personnel regarding the taking of precautionary measures in case the threat level seems to adopt an upward trajectory, even before it reaches the highest level. In addition, its probabilistic basis allows for taking into account noisy data. The applicability of the proposed framework is showcased in a use case that focuses on the identification of potential threats in public events on the basis of evidence obtained from the automatic visual analysis of the footage of surveillance cameras.

show abstract

“…Video image processing to estimate the density distribution of crowd gathering areas is the key to crowd counting. Video crowd counting has important application value in the fields of traffic management, disaster prevention, and public administration [1][2][3][4][5][6]. When the number of crowded people reaches the safety limit, pedestrians are reminded to evacuate safely from the enclosed area to escape the danger, so estimating the number of crowds is helpful for early warning.…”

Section: Introductionmentioning

confidence: 99%

Research on Local Counting and Object Detection of Multiscale Crowds in Video Based on Time-Frequency Analysis

Ren

2022

Journal of Sensors

View full text Add to dashboard Cite

Objective. It has become a very difficult task for cameras to complete real-time crowd counting under congestion conditions. Methods. This paper proposes a DRC-ConvLSTM network, which combines a depth-aware model and depth-adaptive Gaussian kernel to extract the spatial-temporal features and depth-level matching of crowd depth space edge constraints in videos, and finally achieves satisfactory crowd density estimation results. The model is trained with weak supervision on a training set of point-labeled images. The design of the detector is to propose a deep adaptive perception network DRD-NET, which can better initialize the size and position of the head detection frame in the image with the help of density map and RGBD-adaptive perception network. Results. The results show that our method achieves the best performance in RGBD dense video crowd counting on five labeled sequence datasets; the MICC dataset, CrowdFlow dataset, FDST dataset, Mall dataset, and UCSD dataset were evaluated to verify its effectiveness. Conclusion. The experimental results show that the proposed DRD-NET model combined with DRC-ConvLSTM outperforms the existing video crowd counting ConvLSTM model, and the effectiveness of the parameters of each part of the model is further proved by ablation experiments.

show abstract

A Crowd Analysis Framework for Detecting Violence Scenes

Cited by 21 publications

References 32 publications

Beyond Tracking in Crowd: Analyzing Crowd based on Physical Characteristics

Beyond Tracking in Crowd: Analyzing Crowd based on Physical Characteristics

Real‐time threat assessment based on hidden Markov models

Research on Local Counting and Object Detection of Multiscale Crowds in Video Based on Time-Frequency Analysis

Contact Info

Product

Resources

About