Detecting and Tracking Small and Dense Moving Objects in Satellite Videos: A Benchmark

TL;DR

This paper introduces a large-scale satellite video dataset for moving object detection and tracking, proposes a motion modeling baseline, and establishes a public benchmark to evaluate various approaches in this challenging domain.

Contribution

The paper provides the first comprehensive satellite video dataset with annotations, a novel motion modeling baseline, and a public benchmark for moving object detection and tracking.

Findings

The dataset contains over 1.6 million object instances and 3,711 trajectories.

The motion modeling baseline improves detection accuracy and reduces false alarms.

Extensive evaluation of existing methods establishes baseline performance in satellite video analysis.

Abstract

Satellite video cameras can provide continuous observation for a large-scale area, which is important for many remote sensing applications. However, achieving moving object detection and tracking in satellite videos remains challenging due to the insufficient appearance information of objects and lack of high-quality datasets. In this paper, we first build a large-scale satellite video dataset with rich annotations for the task of moving object detection and tracking. This dataset is collected by the Jilin-1 satellite constellation and composed of 47 high-quality videos with 1,646,038 instances of interest for object detection and 3,711 trajectories for object tracking. We then introduce a motion modeling baseline to improve the detection rate and reduce false alarms based on accumulative multi-frame differencing and robust matrix completion. Finally, we establish the first public benchmark for moving object detection and tracking in satellite videos, and extensively evaluate the performance of several representative approaches on our dataset. Comprehensive experimental analyses and insightful conclusions are also provided. The dataset is available at https://github.com/QingyongHu/VISO.