TUMTraf EMOT: Event-Based Multi-Object Tracking Dataset and Baseline for Traffic Scenarios

TL;DR

This paper introduces TUMTraf EMOT, a novel event-based dataset for multi-object tracking in traffic scenarios, addressing limitations of traditional cameras under challenging conditions and establishing a baseline benchmark.

Contribution

The paper presents the first dedicated event-based traffic dataset and a tracking benchmark, advancing research in event-based intelligent transportation systems.

Findings

High performance tracking baseline established

Event cameras outperform traditional cameras in low-light conditions

Dataset enables future research in event-based traffic monitoring

Abstract

In Intelligent Transportation Systems (ITS), multi-object tracking is primarily based on frame-based cameras. However, these cameras tend to perform poorly under dim lighting and high-speed motion conditions. Event cameras, characterized by low latency, high dynamic range and high temporal resolution, have considerable potential to mitigate these issues. Compared to frame-based vision, there are far fewer studies on event-based vision. To address this research gap, we introduce an initial pilot dataset tailored for event-based ITS, covering vehicle and pedestrian detection and tracking. We establish a tracking-by-detection benchmark with a specialized feature extractor based on this dataset, achieving excellent performance.