An Attribute-based Method for Video Anomaly Detection

Code

• github.com/talreiss/accurate-interpretable-vad
  OfficialIn paperpytorch★ 106
• github.com/talreiss/Mean-Shifted-Anomaly-Detection
  pytorch★ 117
• github.com/talreiss/PANDA
  pytorch★ 96
• github.com/openvinotoolkit/anomalib/tree/main/src/anomalib/models/ai_vad
  pytorch★ 0

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.

Tasks

Benchmark Results

Reproductions