Training-Free Time-Series Anomaly Detection: Leveraging Image Foundation Models

TL;DR

This paper introduces a training-free, image-based method for time-series anomaly detection that leverages image foundation models, avoiding unstable training and hyperparameter tuning, while effectively identifying anomalies across frequencies.

Contribution

It proposes a novel approach converting time-series into images for anomaly detection using foundation models, eliminating the need for training deep neural networks.

Findings

Achieves high anomaly detection performance without training neural networks.

Effectively detects anomalies across different frequency components.

Outperforms or matches deep learning models on benchmark datasets.

Abstract

Recent advancements in time-series anomaly detection have relied on deep learning models to handle the diverse behaviors of time-series data. However, these models often suffer from unstable training and require extensive hyperparameter tuning, leading to practical limitations. Although foundation models present a potential solution, their use in time series is limited. To overcome these issues, we propose an innovative image-based, training-free time-series anomaly detection (ITF-TAD) approach. ITF-TAD converts time-series data into images using wavelet transform and compresses them into a single representation, leveraging image foundation models for anomaly detection. This approach achieves high-performance anomaly detection without unstable neural network training or hyperparameter tuning. Furthermore, ITF-TAD identifies anomalies across different frequencies, providing users with a detailed visualization of anomalies and their corresponding frequencies. Comprehensive experiments on five benchmark datasets, including univariate and multivariate time series, demonstrate that ITF-TAD offers a practical and effective solution with performance exceeding or comparable to that of deep models.