arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.16652cs.CVcs.AI
收藏

面向实时无人机跟踪的渐进式表示学习

Progressive Representation Learning for Real-Time UAV Tracking

Tongji University(同济大学);the University of Hong Kong(香港大学)

Changhong Fu, Xiang Lei, Haobo Zuo, Liangliang Yao, Guangze Zheng, Jia Pan

AI总结:

本文提出PRL-Track框架,通过粗粒度与细粒度表示学习解决无人机跟踪中的长宽比变化和遮挡问题,在边缘设备上实现42.6帧每秒的实时高性能跟踪。

摘要:

视觉目标跟踪极大地推动了无人机(UAVs)的自主应用。然而,在复杂动态环境中为无人机跟踪学习鲁棒的目标表示极具挑战性,尤其是在面临长宽比变化和遮挡时。这些挑战严重改变了目标的原始信息。为解决上述问题,本文提出了一种新颖的面向无人机跟踪的渐进式表示学习框架,即PRL-Track。具体而言,PRL-Track分为粗粒度表示学习和细粒度表示学习。对于粗粒度表示学习,设计了两个依赖外观和语义信息的创新调节器,以减轻外观干扰并捕获语义信息。此外,对于细粒度表示学习,开发了一种新的分层建模生成器来交织粗粒度目标表示。详尽的实验表明,所提出的PRL-Track在三个权威无人机跟踪基准上表现出色。真实世界测试表明,在配备边缘智能相机的典型无人机平台上,所提出的PRL-Track实现了卓越的跟踪性能,速度达到42.6帧每秒。代码、模型和演示视频可在\url{https://github.com/vision4robotics/PRL-Track}获取。

英文摘要:

Visual object tracking has significantly promoted autonomous applications for unmanned aerial vehicles (UAVs). However, learning robust object representations for UAV tracking is especially challenging in complex dynamic environments, when confronted with aspect ratio change and occlusion. These challenges severely alter the original information of the object. To handle the above issues, this work proposes a novel progressive representation learning framework for UAV tracking, i.e., PRL-Track. Specifically, PRL-Track is divided into coarse representation learning and fine representation learning. For coarse representation learning, two innovative regulators, which rely on appearance and semantic information, are designed to mitigate appearance interference and capture semantic information. Furthermore, for fine representation learning, a new hierarchical modeling generator is developed to intertwine coarse object representations. Exhaustive experiments demonstrate that the proposed PRL-Track delivers exceptional performance on three authoritative UAV tracking benchmarks. Real-world tests indicate that the proposed PRL-Track realizes superior tracking performance with 42.6 frames per second on the typical UAV platform equipped with an edge smart camera. The code, model, and demo videos are available at \url{https://github.com/vision4robotics/PRL-Track}.

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。

↑