An Attribute-based Method for Video Anomaly Detection

TL;DR

This paper introduces a simple attribute-based video anomaly detection method that achieves state-of-the-art results by representing objects through velocity and pose, and combining these with pretrained deep features.

Contribution

The paper presents a novel, simple attribute-based approach for VAD that outperforms existing methods on standard datasets, using minimal representations.

Findings

Achieves 99.1% AUROC on Ped2

Achieves 93.7% AUROC on Avenue

Achieves 85.9% AUROC on ShanghaiTech

Abstract

Video anomaly detection (VAD) identifies suspicious events in videos, which is critical for crime prevention and homeland security. In this paper, we propose a simple but highly effective VAD method that relies on attribute-based representations. The base version of our method represents every object by its velocity and pose, and computes anomaly scores by density estimation. Surprisingly, this simple representation is sufficient to achieve state-of-the-art performance in ShanghaiTech, the most commonly used VAD dataset. Combining our attribute-based representations with an off-the-shelf, pretrained deep representation yields state-of-the-art performance with a $99.1\%, 93.7\%$, and $85.9\%$ AUROC on Ped2, Avenue, and ShanghaiTech, respectively.