Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectors

TL;DR

This paper introduces a lightweight, self-distilled masked auto-encoder for efficient video anomaly detection, focusing on foreground motion, using dual decoders, and synthetic abnormal data to achieve high speed and competitive accuracy.

Contribution

The paper presents a novel, efficient video anomaly detection model that incorporates motion-gradient-based token weighting, a teacher-student decoder architecture, and synthetic abnormal event generation.

Findings

Achieves 1655 FPS processing speed.

Outperforms or matches state-of-the-art accuracy on four benchmarks.

Demonstrates significant speed improvement over existing methods.

Abstract

We propose an efficient abnormal event detection model based on a lightweight masked auto-encoder (AE) applied at the video frame level. The novelty of the proposed model is threefold. First, we introduce an approach to weight tokens based on motion gradients, thus shifting the focus from the static background scene to the foreground objects. Second, we integrate a teacher decoder and a student decoder into our architecture, leveraging the discrepancy between the outputs given by the two decoders to improve anomaly detection. Third, we generate synthetic abnormal events to augment the training videos, and task the masked AE model to jointly reconstruct the original frames (without anomalies) and the corresponding pixel-level anomaly maps. Our design leads to an efficient and effective model, as demonstrated by the extensive experiments carried out on four benchmarks: Avenue, ShanghaiTech, UBnormal and UCSD Ped2. The empirical results show that our model achieves an excellent trade-off between speed and accuracy, obtaining competitive AUC scores, while processing 1655 FPS. Hence, our model is between 8 and 70 times faster than competing methods. We also conduct an ablation study to justify our design. Our code is freely available at: https://github.com/ristea/aed-mae.