Violence Recognition

Violence recognition in video aims to automatically identify violent acts from visual data, primarily focusing on improving accuracy and efficiency. Current research emphasizes deep learning approaches, employing advanced architectures like vision transformers and incorporating techniques such as 3D skeleton point cloud analysis and salient image representation to enhance performance. These advancements are driven by the need for robust and scalable solutions for applications such as public safety and security, particularly in analyzing large volumes of surveillance footage. The development of specialized datasets and the ongoing refinement of model architectures are key factors in advancing this field.

Papers

October 4, 2023

Reinforcement Learning-based Mixture of Vision Transformers for Video Violence Recognition
Hamid Mohammadi, Ehsan Nazerfard, Tahereh Firoozi
Vision Transformer Positive Reinforcement Learning Mixture Violence Recognition

August 26, 2023

Improving Video Violence Recognition with Human Interaction Learning on 3D Skeleton Point Clouds
Yukun Su, Guosheng Lin, Qingyao Wu
Skeleton Sequence Video Action Recognition Skeleton Based Human Human Interaction Violence Recognition

July 26, 2022

SSIVD-Net: A Novel Salient Super Image Classification & Detection Technique for Weaponized Violence
Toluwani Aremu, Li Zhiyuan, Reem Alameeri, Mustaqeem Khan, Abdulmotaleb El Saddik
MAESTRO Dataset Detection Method Salient Object Detection Surveillance Video Violence Detection Communal Violence Violence Recognition

Violence Recognition

Papers

Reinforcement Learning-based Mixture of Vision Transformers for Video Violence Recognition

Improving Video Violence Recognition with Human Interaction Learning on 3D Skeleton Point Clouds

SSIVD-Net: A Novel Salient Super Image Classification & Detection Technique for Weaponized Violence