InDiD: Instant Disorder Detection via Representation Learning

TL;DR

This paper introduces InDiD, a novel deep learning approach with a specialized loss function for rapid and accurate change point detection in various sequential data types, including challenging video data.

Contribution

It presents a differentiable loss function enabling representation learning tailored for change point detection, outperforming traditional methods on synthetic and real-world datasets.

Findings

InDiD achieves higher F1 scores in explosion detection in videos.

The method effectively learns meaningful representations for complex data.

It outperforms baseline change point detection methods.

Abstract

For sequential data, a change point is a moment of abrupt regime switch in data streams. Such changes appear in different scenarios, including simpler data from sensors and more challenging video surveillance data. We need to detect disorders as fast as possible. Classic approaches for change point detection (CPD) might underperform for semi-structured sequential data because they cannot process its structure without a proper representation. We propose a principled loss function that balances change detection delay and time to a false alarm. It approximates classic rigorous solutions but is differentiable and allows representation learning for deep models. We consider synthetic sequences, real-world data sensors and videos with change points. We carefully labelled available data with change point moments for video data and released it for the first time. Experiments suggest that complex data require meaningful representations tailored for the specificity of the CPD task -- and our approach provides them outperforming considered baselines. For example, for explosion detection in video, the F1 score for our method is 0.53 compared to baseline scores of 0.31 and 0.35.