Human Activity Recognition Method for Moderate Violence Detection

TL;DR

This paper presents a real-time computer vision system that detects moderate physical violence, such as pushing, in surveillance footage using skeletal analysis and machine learning, achieving high precision even in challenging conditions.

Contribution

The study introduces a novel approach combining human detection, skeletal keypoint extraction, and a Random Forest classifier for violence detection in surveillance videos.

Findings

Achieved 0.98 precision in controlled environments.

Maintained 0.72 precision in challenging real-world scenarios.

Demonstrated feasibility of skeletal analysis for early violence detection.

Abstract

Physical violence in public spaces is a significant public health concern, with minor incidents such as pushing often serving as precursors to more severe escalations. This research develops an automated system for the real-time detection of moderate physical violence, specifically pushing, in surveillance camera footage. The proposed solution integrates state-of-the-art computer vision models, utilizing YOLO11 and YOLO11-Pose for human detection and skeletal keypoint extraction. By calculating body inclination and joint angles between shoulders and hips, a Random Forest classifier was trained to distinguish between normal behavior and aggressive physical contact. The system's performance was evaluated through three progressive case studies representing increasing levels of difficulty. In controlled environments with frontal views, the model achieved a precision of 0.98. In the most challenging scenario, featuring high-altitude, steep-angle recordings from real-world surveillance infrastructure, the system maintained a precision of 0.72 despite significant perspective distortion and visual noise. These results demonstrate the feasibility of using skeletal analysis for early violence intervention in urban security contexts.