Toward Fast and Accurate Violence Detection for Automated Video Surveillance Applications

Surveillance cameras are increasingly utilized worldwide, but the large volume of generated video data poses challenges for real-time monitoring. This paper presents a deep learning-based approach for automatic violence detection in surveillance footage. By leveraging 3D convolutional neural networks and transfer learning from pre-trained action recognition models, we propose an efficient and accurate system for identifying violent activities. Our experimental evaluation on multiple datasets demonstrates improved classification accuracy with fewer model parameters compared to state-of-the-art methods.