Multiple Object Tracking in Satellite Video With Graph-Based Multi-Clue Fusion Tracker

With the rapid advancement of satellite technology, satellite video has emerged as a key method for acquiring dynamic terrestrial information, facilitating the multiple object tracking (MOT). Satellites are capable of surveying vast urban landscapes, yet the observed objects are small and dispersed...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on geoscience and remote sensing 2024-09, p.1-1
Main Authors: Chen, Haoxiang, Li, Nannan, Li, Dongjin, Lv, Jianwei, Zhao, Wei, Zhang, Rufei, Xu, Jingyu
Format: Article
Language:English
Subjects:
Feature extraction
graph neural networks
Image edge detection
Interference
Location awareness
multiple object tracking (MOT)
relational feature
Satellite video
Satellites
Target tracking
task alignment
tiny object detection (TOD)
Trajectory
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the rapid advancement of satellite technology, satellite video has emerged as a key method for acquiring dynamic terrestrial information, facilitating the multiple object tracking (MOT). Satellites are capable of surveying vast urban landscapes, yet the observed objects are small and dispersed among complex interference from background, heightening the challenges in detection and association tasks for object tracking. However, current trackers often dissociate the classification task from the localization task, leading to drift in tiny object detection, and rely on prior knowledge for clue ranking, limiting model robustness. In this paper, we introduce the Graph-based Multi-clue Fusion Tracker (GMFTracker). Initially, we introduce a sparse sampling based feature map correction approach to rectify the misalignment between the classification and localization feature maps. Furthermore, we developed a graph neural networks (GNNs) for object relationship modeling, free from presuppositions, to tackle association challenges using relational feature. GMFTracker was rigorously tested on VISO, CGSTL and TinyPerson datasets, demonstrating its competitive performance relative to contemporary studies.
ISSN:0196-2892
1558-0644
DOI:10.1109/TGRS.2024.3457517