arXiv 2601.092402026-01cs.CVeess.IV

DeTracker：在未稳定卫星视频中实现运动解耦的车辆检测与跟踪

DeTracker: Motion-decoupled Vehicle Detection and Tracking in Unstabilized Satellite Videos

State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University（武汉大学测绘遥感信息工程国家重点实验室）；School of Artificial Intelligence, Wuhan University（武汉大学人工智能学院）；Hangzhou Institute of Technology, Xidian University（西安电子科技大学杭州研究院）

Jiajun Chen, Jing Xiao, Shaohan Cao, Yuming Zhu, Liang Liao, Jun Pan, Mi Wang

发表信息：: IEEE Transactions on Geoscience and Remote Sensing, vol. 64, Art. no. 5623214, 2026
DOI：: 10.1109/TGRS.2026.3694908

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对未稳定卫星视频中多目标跟踪的挑战，DeTracker提出一种联合检测与跟踪框架，通过全局局部运动解耦模块和时间依赖特征金字塔模块提升轨迹稳定性与目标识别一致性，实验表明其在SDM-Car-SU和真实卫星视频数据上均取得优异性能。

摘要：

卫星视频提供连续的地表动态观测，但对多目标跟踪（MOT）构成重大挑战，特别是在未稳定条件下，平台抖动和微小目标弱外观共同劣化跟踪性能。为解决此问题，我们提出DeTracker，一种针对未稳定卫星视频的联合检测与跟踪框架。DeTracker引入任务驱动的全局-局部运动解耦（GLMD）模块，以解决主导平台运动与弱目标运动之间的运动不平衡。它通过特征层面的全局语义对齐抑制背景主导的运动，并通过局部细化捕捉目标特定的运动，从而提高轨迹稳定性和身份一致性。此外，开发了时间依赖特征金字塔（TDFP）模块，用于跨帧时间特征融合，增强微小目标表示的连续性和判别性。我们进一步构建了一个新的基准数据集SDM-Car-SU，模拟多方向和多速度平台运动，以系统评估跟踪鲁棒性在不同运动扰动下的表现。在模拟和真实未稳定卫星视频上的大量实验表明，DeTracker显著优于现有方法，在SDM-Car-SU上达到61.1%的MOTA，在真实卫星视频数据上达到45.3%的MOTA。代码和数据集将在https://github.com/alex-chenjiajun/DeTracker上公开。

英文摘要：

Satellite videos provide continuous observations of surface dynamics but pose significant challenges for multi-object tracking (MOT), especially under unstabilized conditions where platform jitter and the weak appearance of tiny objects jointly degrade tracking performance. To address this problem, we propose DeTracker, a joint-detection-and-tracking framework tailored for unstabilized satellite videos. DeTracker introduces a task-driven Global-Local Motion Decoupling (GLMD) module to address the motion imbalance between dominant platform motion and weak target motion. It suppresses background-dominated motion via global semantic alignment at the feature level and captures target-specific motion through local refinement, improving trajectory stability and identity consistency. In addition, a Temporal Dependency Feature Pyramid (TDFP) module is developed to perform cross-frame temporal feature fusion, enhancing the continuity and discriminability of tiny-object representations. We further construct a new benchmark dataset, SDM-Car-SU, which simulates multi-directional and multi-speed platform motions to enable systematic evaluation of tracking robustness under varying motion perturbations. Extensive experiments on both simulated and real unstabilized satellite videos demonstrate that DeTracker significantly outperforms existing methods, achieving 61.1% MOTA on SDM-Car-SU and 45.3% MOTA on real satellite video data. The code and dataset will be publicly available at https://github.com/alex-chenjiajun/DeTracker.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。