Grid HTM: Hierarchical Temporal Memory for Anomaly Detection in Videos

TL;DR

This paper introduces Grid HTM, a novel Hierarchical Temporal Memory architecture designed for anomaly detection in complex videos, addressing limitations of deep learning methods like noise sensitivity and lack of online learning.

Contribution

The paper presents Grid HTM, a new HTM-based architecture tailored for video anomaly detection, offering advantages over traditional deep learning approaches.

Findings

Grid HTM demonstrates noise tolerance in video analysis.

It supports online learning for real-time anomaly detection.

Effective in complex surveillance scenarios.

Abstract

The interest for video anomaly detection systems has gained traction for the past few years. The current approaches use deep learning to perform anomaly detection in videos, but this approach has multiple problems. For starters, deep learning in general has issues with noise, concept drift, explainability, and training data volumes. Additionally, anomaly detection in itself is a complex task and faces challenges such as unknowness, heterogeneity, and class imbalance. Anomaly detection using deep learning is therefore mainly constrained to generative models such as generative adversarial networks and autoencoders due to their unsupervised nature, but even they suffer from general deep learning issues and are hard to train properly. In this paper, we explore the capabilities of the Hierarchical Temporal Memory (HTM) algorithm to perform anomaly detection in videos, as it has favorable properties such as noise tolerance and online learning which combats concept drift. We introduce a novel version of HTM, namely, Grid HTM, which is an HTM-based architecture specifically for anomaly detection in complex videos such as surveillance footage.