提出边缘感知热红外无人机群跟踪方法，融合图谱卷积与注意力机制，突破多目标遮挡和红外模糊难题。
arXiv:2607.12544v1 Announce Type: new Abstract: Thermal infrared (TIR) imaging is essential for UAV swarm operations in visually degraded environments…
诊断Video-LLM在长视频中跟踪角色的失败，揭示当前最强开源模型未能真正“观看”视频的局限。
arXiv:2607.11078v1 Announce Type: cross Abstract: Can a Video Large Language Model (Video-LLM) follow one person through a long video, keeping track o…
将几何信息融入目标跟踪模型，G²TAM在Track Anything基础上显著提升鲁棒性与精度，ICML 2026收录的新范式。
arXiv:2607.03789v1 Announce Type: new Abstract: Human spatial understanding arises from jointly perceiving geometry and semantics, enabling consistent…
多目标跟踪迎来新世代！GenTrack提出生成式架构，在复杂场景下实现更精准的轨迹预测与关联。
arXiv:2510.24399v2 Announce Type: replace Abstract: This paper introduces a novel multi-object tracking (MOT) method, dubbed GenTrack, whose main cont…
最新TPAMI论文：利用文本到图像扩散模型实现无监督视觉目标跟踪，无需标注数据即可追踪目标，开辟扩散模型新应用。
arXiv:2605.26933v1 Announce Type: new Abstract: Unsupervised visual object tracking is a challenging task that requires following arbitrary targets in…