Object-centric and memory-guided normality reconstruction for video anomaly detection

TL;DR

This paper proposes a novel object-centric, memory-guided approach for video anomaly detection that models normal patterns without using anomalous data, combining appearance, motion, and prototype memory for improved accuracy.

Contribution

It introduces a method coupling pretrained object-level features with a cosine distance anomaly measure, enhancing reconstruction-based models with additional constraints.

Findings

Outperforms state-of-the-art on multiple datasets

Effectively models normal object behaviors without anomalous training data

Utilizes combined appearance and motion features for better detection

Abstract

This paper addresses video anomaly detection problem for videosurveillance. Due to the inherent rarity and heterogeneity of abnormal events, the problem is viewed as a normality modeling strategy, in which our model learns object-centric normal patterns without seeing anomalous samples during training. The main contributions consist in coupling pretrained object-level action features prototypes with a cosine distance-based anomaly estimation function, therefore extending previous methods by introducing additional constraints to the mainstream reconstruction-based strategy. Our framework leverages both appearance and motion information to learn object-level behavior and captures prototypical patterns within a memory module. Experiments on several well-known datasets demonstrate the effectiveness of our method as it outperforms current state-of-the-art on most relevant spatio-temporal evaluation metrics.