Multiple Pedestrians and Vehicles Tracking in Aerial Imagery: A Comprehensive Study

TL;DR

This paper presents AerialMPTNet, a novel deep learning multi-object tracking method for aerial imagery that combines appearance, temporal, and graphical data, outperforming previous approaches on multiple datasets.

Contribution

Introduction of AerialMPTNet, a multi-object tracking model integrating Siamese, LSTM, and GCN modules, with novel use of SE layers and OHEM in regression-based tracking.

Findings

AerialMPTNet outperforms previous methods on pedestrian datasets.

LSTM and GCN modules improve tracking accuracy.

L1 loss generally yields better results than Huber loss.

Abstract

In this paper, we address various challenges in multi-pedestrian and vehicle tracking in high-resolution aerial imagery by intensive evaluation of a number of traditional and Deep Learning based Single- and Multi-Object Tracking methods. We also describe our proposed Deep Learning based Multi-Object Tracking method AerialMPTNet that fuses appearance, temporal, and graphical information using a Siamese Neural Network, a Long Short-Term Memory, and a Graph Convolutional Neural Network module for a more accurate and stable tracking. Moreover, we investigate the influence of the Squeeze-and-Excitation layers and Online Hard Example Mining on the performance of AerialMPTNet. To the best of our knowledge, we are the first in using these two for a regression-based Multi-Object Tracking. Additionally, we studied and compared the L1 and Huber loss functions. In our experiments, we extensively evaluate AerialMPTNet on three aerial Multi-Object Tracking datasets, namely AerialMPT and KIT AIS pedestrian and vehicle datasets. Qualitative and quantitative results show that AerialMPTNet outperforms all previous methods for the pedestrian datasets and achieves competitive results for the vehicle dataset. In addition, Long Short-Term Memory and Graph Convolutional Neural Network modules enhance the tracking performance. Moreover, using Squeeze-and-Excitation and Online Hard Example Mining significantly helps for some cases while degrades the results for other cases. In addition, according to the results, L1 yields better results with respect to Huber loss for most of the scenarios. The presented results provide a deep insight into challenges and opportunities of the aerial Multi-Object Tracking domain, paving the way for future research.