VelocityNet: Real-Time Crowd Anomaly Detection via Person-Specific Velocity Analysis

TL;DR

VelocityNet is a real-time crowd anomaly detection framework that leverages person-specific velocity analysis and hierarchical clustering to identify abnormal motion patterns in crowded scenes.

Contribution

The paper introduces VelocityNet, a novel dual-pipeline approach combining head detection and optical flow for interpretable, real-time crowd anomaly detection.

Findings

Effective in real-time detection of diverse anomalies

Handles high crowd densities with occlusions

Provides interpretable anomaly indicators

Abstract

Detecting anomalies in crowded scenes is challenging due to severe inter-person occlusions and highly dynamic, context-dependent motion patterns. Existing approaches often struggle to adapt to varying crowd densities and lack interpretable anomaly indicators. To address these limitations, we introduce VelocityNet, a dual-pipeline framework that combines head detection and dense optical flow to extract person-specific velocities. Hierarchical clustering categorizes these velocities into semantic motion classes (halt, slow, normal, and fast), and a percentile-based anomaly scoring system measures deviations from learned normal patterns. Experiments demonstrate the effectiveness of our framework in real-time detection of diverse anomalous motion patterns within densely crowded environments.