Deep Learning in Video Violence Detection

Su, Minglan; Zhang, Chaoying; Tong, Ying; Liang, Baolin; Ma, Sicong; Wang, Jianxiu

doi:10.1109/ctmcd53128.2021.00064

Cited by 8 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Paper [13], The study uses Hockey Fights and Movies Dataset. This contains 500 violent and nonviolent each.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Brutality detection and rendering of brutal frames

Madana Mohana,

Singh,

Kumar

et al. 2024

MATEC Web Conf.

View full text Add to dashboard Cite

The popularity of anime is increasing exponentially in every part of the world due to its unique storyline, nonstop entertainment, fights, and similar type of content that can hold viewers and keeps them at the edge of their seats. However, with the increase of popularity in anime there has also been an exponential increase in violence and brutality in anime videos. Violent scenes have become much more common in anime videos when compared to generic cinema. This survey paper presents a comprehensive view on the detection of violence in movies and different scenarios using various techniques. Most commonly to automate detection of violence, machine learning is used for training the machine to detect violence. Convolution neural networks (CNN) are used very commonly to understand image pattern recognition with high accuracy. Moreover, use of other different methods such as LSTM and Markov models are also used to detect violence. The main goals kept in mind while working is to detect violence with high accuracy and to use less computation or to perform the action at a high-speed rate.

show abstract

“…In Paper [13], The study uses Hockey Fights and Movies Dataset. This contains 500 violent and nonviolent each.…”

Section: Literature Reviewmentioning

confidence: 99%

“…In Paper [15], Datasets: Hockey Fights dataset is 121.271 seconds using DWT-SVM with a polynomial kernel. The polynomial kernel testing is the fastest.…”

Section: Literature Reviewmentioning

confidence: 99%

Brutality detection and rendering of brutal frames

Madana Mohana,

Singh,

Kumar

et al. 2024

MATEC Web Conf.

View full text Add to dashboard Cite

show abstract

“…Regarding the study of groups of people, advances in analysing behaviour are limited to very concrete and simple activities or actions, usually of short duration (low semantic component) such as a actions in sport games [13], [37], [42], [53], detection interactions of people inside a group [15], [57], [60], inter-group violence [51], [64], [65], among others. If we increase the number of people in the group, becoming crowds, the level of semantics is even lower, being specifically limited to tasks such as counting people and calculating crowd density [8], [18], [25], [68] or detecting movements of a mass of people or crowd collisions [21], [39], [49], [71], mainly for the purpose of security tasks.…”

Section: Related Workmentioning

confidence: 99%

“…For example, the use of Convolutional Neural Networks (CNN) is shown in the work of Li et al [29] where a new colorization of images including other information as optical flow is used as input of CNNs to detect objects and their anomalies. Also, in [51], Su et al integrate the one-class Support Vector Machine into a CNN proposing the Deep One-Class (DOC) model. One widely used model has been autoencoders (AE) where they attempt to extract features from images to form a new space in which to decide the existence of normal activities.…”

Section: Related Workmentioning

confidence: 99%

Architecture for Automatic Recognition of Group Activities Using Local Motions and Context

et al. 2022

View full text Add to dashboard Cite

Currently, the ability to automatically detect human behavior in image sequences is one of the most important challenges in the area of computer vision. Within this broad field of knowledge, the recognition of activities of people groups in public areas is receiving special attention due to its importance in many aspects including safety and security. This paper proposes a generic computer vision architecture with the ability to learn and recognize different group activities using mainly the local group's movements. Specifically, a multi-stream deep learning architecture is proposed whose two main streams correspond to a representation based on a descriptor capable of representing the trajectory information of a sequence of images as a collection of local movements that occur in specific regions of the scene. Additional information (e.g. location, time, etc.) to strengthen the classification of activities by including it as additional streams. The proposed architecture is capable of classifying in a robust way different activities of a group as well to deal with the one-class problems. Moreover, the use of a simple descriptor that transforms a sequence of color images into a sequence of two-image streams can reduce the curse of dimensionality using a deep learning approach. The generic deep learning architecture has been evaluated with different datasets outperforming the state-of-the-art approaches providing an efficient architecture for single and multi-class classification problems.

show abstract

“…Both [14] and [15] provided a comprehensive survey of deep learning-based models for video anomaly detection with minor differences in classifications, while [14] has an additional part evaluating performance among the models. Su et al [16] summarised the latest methods of violence detection in existing video sequences. Roshan et al [17] reviewed the recent trends in violence detection, and performed a comparative study of different state-of-the-art shallow and deep models.…”

Section: A Relevant Surveysmentioning

confidence: 99%

Deep Video Anomaly Detection: Opportunities and Challenges

Ren¹,

Xia²,

Liu³

et al. 2021

2021 International Conference on Data Mining Workshops (ICDMW)

View full text Add to dashboard Cite

Anomaly detection is a popular and vital task in various research contexts, which has been studied for several decades. To ensure the safety of people's lives and assets, video surveillance has been widely deployed in various public spaces, such as crossroads, elevators, hospitals, banks, and even in private homes. Deep learning has shown its capacity in a number of domains, ranging from acoustics, images, to natural language processing. However, it is non-trivial to devise intelligent video anomaly detection systems cause anomalies significantly differ from each other in different application scenarios. There are numerous advantages if such intelligent systems could be realised in our daily lives, such as saving human resources in a large degree, reducing financial burden on the government, and identifying the anomalous behaviours timely and accurately. Recently, many studies on extending deep learning models for solving anomaly detection problems have emerged, resulting in beneficial advances in deep video anomaly detection techniques. In this paper, we present a comprehensive review of deep learning-based methods to detect the video anomalies from a new perspective. Specifically, we summarise the opportunities and challenges of deep learning models on video anomaly detection tasks, respectively. We put forth several potential future research directions of intelligent video anomaly detection system in various application domains. Moreover, we summarise the characteristics and technical problems in current deep learning methods for video anomaly detection.

show abstract

Deep Learning in Video Violence Detection

Cited by 8 publications

References 6 publications

Brutality detection and rendering of brutal frames

Brutality detection and rendering of brutal frames

Architecture for Automatic Recognition of Group Activities Using Local Motions and Context

Deep Video Anomaly Detection: Opportunities and Challenges

Contact Info

Product

Resources

About