Anomalous Motion Detection on Highway Using Deep Learning

TL;DR

This paper introduces the Highway Traffic Anomaly (HTA) dataset for detecting unusual traffic patterns from dash cam videos and evaluates deep learning models, revealing challenges and proposing improvements for dynamic environments.

Contribution

The paper presents a new dataset for highway anomaly detection and adapts deep learning models to better handle dynamic traffic scenarios.

Findings

State-of-the-art models perform poorly on dynamic highway data

Proposed model variations improve anomaly detection accuracy

New dataset enables research in real-world traffic anomaly detection

Abstract

Research in visual anomaly detection draws much interest due to its applications in surveillance. Common datasets for evaluation are constructed using a stationary camera overlooking a region of interest. Previous research has shown promising results in detecting spatial as well as temporal anomalies in these settings. The advent of self-driving cars provides an opportunity to apply visual anomaly detection in a more dynamic application yet no dataset exists in this type of environment. This paper presents a new anomaly detection dataset - the Highway Traffic Anomaly (HTA) dataset - for the problem of detecting anomalous traffic patterns from dash cam videos of vehicles on highways. We evaluate state-of-the-art deep learning anomaly detection models and propose novel variations to these methods. Our results show that state-of-the-art models built for settings with a stationary camera do not translate well to a more dynamic environment. The proposed variations to these SoTA methods show promising results on the new HTA dataset.