Related references
Note: Only part of the references are listed.
Article
Computer Science, Information Systems
Peng Wu et al.
Summary: Violence detection in videos has great potential in practical applications. Previous works have treated it as a simple video classification task, but this approach has limitations. To address this, we propose a new approach that uses large-scale audio-visual violence data and introduces two complementary tasks, namely coarse-grained violent frame detection and fine-grained violent event detection. Our approach outperforms state-of-the-art methods on public datasets, and the use of audio-visual input and relationship modeling has a positive effect.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Artificial Intelligence
Momina Masood et al.
Summary: Easy access to audio-visual content on social media and the availability of modern tools and open-source models have led to the emergence of deepfakes, which can be used for deception, spreading disinformation, revenge porn, financial frauds, and disruption of government functioning. This paper provides a comprehensive review of existing tools and machine learning approaches for deepfake generation and detection, discussing manipulation approaches, datasets, performance evaluation standards, and future directions.
APPLIED INTELLIGENCE
(2023)
Article
Women's Studies
Diarmaid Harkin et al.
Summary: This article points out that technology-based responses are necessary but not sufficient in addressing TFA, and they have persistent limitations. Therefore, there is a need for continued development of human resources, especially professional domestic violence support workers, to provide support and assistance to those experiencing TFA.
VIOLENCE AGAINST WOMEN
(2023)
Article
Computer Science, Artificial Intelligence
Gianni Brauwers et al.
Summary: Attention is a vital mechanism in various domains and tasks of deep learning. This survey offers an overview of the important attention mechanisms proposed in literature. The mechanisms are explained using a framework comprising a general attention model, uniform notation, and a comprehensive taxonomy. Additionally, it reviews evaluation measures and methods to characterize the structure of attention models based on the framework. Future work in the field of attention models is also considered.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Review
Psychology, Multidisciplinary
Tania Johnston et al.
Summary: Virtual embodiment has been shown to have a significant impact on reducing intimate partner violence (IPV) by enhancing emotional recognition and decreasing violent behaviors, through the integration of neuroscience-inspired and validated state-of-the-art technology.
FRONTIERS IN PSYCHOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Peng Xu et al.
Summary: This paper presents a comprehensive survey of Transformer techniques in multimodal learning, covering different variants of Transformers, applications in multimodal pretraining and specific tasks, common challenges and designs, as well as future research directions.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Software Engineering
Seyed Mehdi Mohtavipour et al.
Summary: This paper proposes a novel deep violence detection framework based on specific features derived from handcrafted methods, covering different aspects of violent behavior. A convolutional neural network (CNN) is trained to learn environment patterns, motion patterns, and representative violent actions, achieving better performance in accuracy and processing time compared to state-of-the-art works.
Article
Chemistry, Analytical
Romas Vijeikis et al.
Summary: This paper presents a novel architecture for violence detection from video surveillance cameras. The proposed model utilizes a spatial feature extracting U-Net-like network with MobileNet V2 as an encoder, followed by LSTM for temporal feature extraction and classification. The computational light model achieves good results, with an average accuracy of 0.82 +/- 2% and average precision of 0.81 +/- 3% on a complex real-world security camera footage dataset based on RWF-2000.
Article
Psychology, Multidisciplinary
Nicolas Barnes et al.
Summary: Virtual reality (VR) is a technology that immerses users in different environments through virtual experiences, and can be used to improve empathy skills and train non-violent responses for intimate partner violence (IPV) perpetrators. This article discusses the practical aspects of using virtual reality for rehabilitation in a prison environment.
FRONTIERS IN PSYCHOLOGY
(2022)
Article
Automation & Control Systems
Fath U. Min Ullah et al.
Summary: This article proposes an AI-enabled IIoT-based framework called VD-Net to intelligently detect and analyze violent scenes, improving the accuracy of surveillance systems.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Shreyan Ganguly et al.
Summary: The prime goal of creating synthetic digital data is to generate something very closer to real ones when the original data are scarce. However, the trustworthiness of such digital content is dipping potentially in society owing to malicious users. In order to verify the originality of face images/videos generated by deepfake methods, a model enhanced with visual attention technique is proposed and evaluated on publicly available datasets. The proposed model outperforms state-of-the-art methods on these datasets.
PATTERN ANALYSIS AND APPLICATIONS
(2022)
Article
Criminology & Penology
Kweilin T. Lucas
Summary: Technology-facilitated abuse, especially through the use of deepfake technology, is becoming increasingly prevalent in cases of domestic violence. Deepfake technology allows for the creation of realistic and manipulated videos and audios, which can be used to threaten, blackmail, and abuse victims. Further research is needed to understand the intersection of deepfake technology and domestic abuse, aiming to guide future policy and address issues of privacy, legal action, and prevention.
VICTIMS & OFFENDERS
(2022)
Article
Computer Science, Information Systems
Jose A. Miranda et al.
Summary: This article explores the problem of gender-based violence and the use of technology to achieve gender equality. It proposes an autonomous multimodal system called Bindi that utilizes artificial intelligence to automatically identify violent situations and trigger a protection protocol. The solution integrates modern technologies such as the Internet of Bodies, affective computing, and cyber-physical systems. The research evaluates the solution using a newly collected dataset and provides an analysis of multimodal fusion strategies and audio data processing.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Proceedings Paper
Computer Science, Information Systems
Alexander Heinrich et al.
Summary: Finder networks, especially Apple's Find My network, can pose serious threats to users' privacy and health if abused for stalking. The release of Apple's AirTag has amplified this issue. While Apple provides stalking detection within its ecosystem, Android users are left without similar protection. Researchers reverse-engineered Apple's tracking protection in iOS, developed AirGuard as an Android app to protect against abuse by Apple tracking devices, and conducted a comparative study with Apple's solution.
PROCEEDINGS OF THE 15TH ACM CONFERENCE ON SECURITY AND PRIVACY IN WIRELESS AND MOBILE NETWORKS (WISEC '22)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Victor E. de S. Silva et al.
Summary: This paper proposes a federated learning approach for violence detection in videos. Various pretrained convolutional neural network architectures were evaluated, and the MobileNet architecture showed the best performance in terms of accuracy.
2022 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Maria Gadelkarim et al.
Summary: This paper proposes a framework for detecting and classifying violent events from video streams. It uses 3D Convolutional Neural Networks for violence detection and Inception-v3 model for feature extraction and temporal processing. The paper also discusses the training of models on different data sources and the application of transfer learning.
2022 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN)
(2022)
Article
Computer Science, Artificial Intelligence
Yuval Nirkin et al.
Summary: The proposed method utilizes two networks to detect face swapping and other identity manipulations in single images, achieving state of the art results in detection accuracy.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Proceedings Paper
Computer Science, Information Systems
Jimmy Briggs et al.
Summary: This article presents a general method and open-source Android application for detecting maliciously deployed BLE trackers. Through a small validation study, it demonstrates high precision and recall for most trackers, despite the challenges posed by AirTags.
2022 43RD IEEE SYMPOSIUM ON SECURITY AND PRIVACY WORKSHOPS (SPW 2022)
(2022)
Article
Computer Science, Information Systems
Kevin B. Kwan-Loo et al.
Summary: Regarding safety and security, the paper discusses the problem of felonies and violent crimes and proposes the use of Artificial Intelligence to detect and predict violent behavior. The authors develop an innovative solution model that focuses on pedestrian detection, tracking, pose estimation, and neural networks. They also create a new database for training and testing the model, and achieve a high accuracy rate on this database. The proposed solution model can be extended to classify other types of behavior as well.
Article
Social Sciences, Mathematical Methods
Phan Trinh Ha et al.
Summary: This study develops a machine learning tool called DETECTIPV, using data from questionnaires on intimate partner violence (IPV), to predict IPV from free text. The results demonstrate that DETECTIPV can identify different types of violence and establish its reliability in detecting IPV in various applications.
JOURNAL OF COMPUTATIONAL SOCIAL SCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Mihai Gabriel Constantin et al.
Summary: This article reports on the creation of a publicly available evaluation framework for Violent Scenes Detection (VSD) in Hollywood and YouTube videos. The comprehensive VSD96 dataset is proposed and the crucial components of VSD algorithms are analyzed. The possibility of going beyond state-of-the-art performance through ad-hoc late fusion approach is discussed.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2022)
Article
Chemistry, Analytical
Jose A. Miranda et al.
Summary: Emotion recognition technology is advancing to protect vulnerable populations in dangerous situations such as personal assaults, child or elder abuse, and gender violence. A specialized fear recognition system based on physiological signals has been proposed, utilizing sensors and AI to detect fear emotion and ensure the safety of women.
Article
Computer Science, Artificial Intelligence
Daniel W. Otter et al.
Summary: This article provides a brief introduction to the recent developments in natural language processing driven by deep learning models, discusses their applications and research findings, and offers recommendations for future research directions.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Information Systems
Sakorn Mekruksavanich et al.
Summary: Currently, there is significant interest in research on Human Activity Recognition (HAR) for practical applications like biometric user identification, elderly health monitoring, and surveillance. Deep learning is the most commonly used approach in HAR systems. A novel framework for multi-class wearable user identification based on deep learning models has been proposed and validated with sensory data from wearable devices. The accuracy of 91.77% and 92.43% for basic models CNN and LSTM, respectively, shows promising results for biometric user identification.
Article
Social Issues
Magdalena Diaz Gorfinkiel et al.
Summary: Intimate partner violence remains a critical social issue, and technology is seen as an innovative tool to address it. This article explores the potential of technology, particularly a new device, in protecting female victims through qualitative analysis of expert testimonies.
TECHNOLOGY IN SOCIETY
(2021)
Article
Mathematics
Carlos M. Castorena et al.
Summary: The issue of gender-based violence in Mexico has significantly increased, and efforts have been made to tackle this issue through the use of deep learning neural networks to identify such violence on Twitter. The results of this study show the effectiveness of this approach in detecting gender violence messages on Twitter, with minimal pre-processing of the data set.
Article
Computer Science, Hardware & Architecture
Aleksandr Ometov et al.
Summary: This paper provides an overview of the history and current state of the wearable device market, including a diverse classification and discussion on technologies, architectures, data processing, and market status. It also highlights existing and future solutions to critical challenges in wearable technology.
Article
Computer Science, Artificial Intelligence
Roberto Caldelli et al.
Summary: Deepfakes pose a serious threat in video manipulation, utilizing AI-based technologies to create realistic videos. A new forensic technique based on optical flow fields is proposed to detect fake and original video sequences, showing comparable performance with state-of-the-art methods and superior robustness in more realistic scenarios. The proposed technique can also be combined with frame-based approaches to enhance overall effectiveness.
PATTERN RECOGNITION LETTERS
(2021)
Article
Computer Science, Information Systems
Md. Murad Hossain et al.
Summary: The COVID-19 outbreak in Bangladesh during the summer of 2020 led to preventative measures and restrictions, resulting in increased social problems such as domestic violence. Machine learning algorithms were utilized to predict family violence, with findings indicating a close relationship between family violence and family income level and education level of family members.
APPLIED SYSTEM INNOVATION
(2021)
Review
Computer Science, Information Systems
Dalia Andrea Rodriguez et al.
Summary: Violence against women and children is a global public health issue, with computer science and related technologies being leveraged to develop solutions. Research shows that as the Internet and social media become more prevalent, forms of violence are evolving and require new prevention measures.
Article
Computer Science, Information Systems
Min-Seok Kang et al.
Summary: This paper proposes a novel violence detection pipeline that combines 2D convolutional neural networks with frame grouping to learn spatio-temporal representations in videos, and introduces spatial and temporal attention modules to enhance violence recognition performance. The proposed pipeline outperforms existing methods in both performance and computational complexity, achieving state-of-the-art results on multiple violence datasets.
Article
Computer Science, Artificial Intelligence
Simone Accattoli et al.
APPLIED ARTIFICIAL INTELLIGENCE
(2020)
Article
Computer Science, Information Systems
Yushi Cheng et al.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2020)
Review
Public, Environmental & Occupational Health
Christo El Morr et al.
Article
Instruments & Instrumentation
Princy Randhawa et al.
Proceedings Paper
Acoustics
Bruno Peixoto et al.
2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Jifu Shi et al.
COMMUNICATIONS, SIGNAL PROCESSING, AND SYSTEMS, CSPS 2018, VOL III: SYSTEMS
(2020)
Article
Computer Science, Information Systems
Endang Wahyu Pamungkas et al.
INFORMATION PROCESSING & MANAGEMENT
(2020)
Article
Computer Science, Information Systems
Hector A. Gonzalez et al.
Review
Criminology & Penology
Hadeel Al-Alosi
AGGRESSION AND VIOLENT BEHAVIOR
(2020)
Article
Chemistry, Analytical
Fath U. Min Ullah et al.
Proceedings Paper
Computer Science, Theory & Methods
Jianquan Liu et al.
2019 2ND IEEE CONFERENCE ON MULTIMEDIA INFORMATION PROCESSING AND RETRIEVAL (MIPR 2019)
(2019)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Ahsan Wajahat et al.
2019 2ND INTERNATIONAL CONFERENCE ON COMPUTING, MATHEMATICS AND ENGINEERING TECHNOLOGIES (ICOMET)
(2019)
Review
Computer Science, Information Systems
Kamran Kowsari et al.
Article
Computer Science, Information Systems
Sudha Subramani et al.
Article
Computer Science, Artificial Intelligence
Ismael I. Serrano et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2018)
Article
Engineering, Electrical & Electronic
Neha Dawar et al.
IEEE SENSORS JOURNAL
(2018)
Article
Multidisciplinary Sciences
Peipei Zhou et al.
Article
Computer Science, Information Systems
Sudha Subramani et al.
Proceedings Paper
Computer Science, Theory & Methods
Sudha Subramani et al.
DATABASES THEORY AND APPLICATIONS, ADC 2018
(2018)
