MV-TAP：在多视角视频中跟踪任意点
MV-TAP: Tracking Any Point in Multi-View Videos
把单目点跟踪扩展到多相机系统，用相机几何加跨视角注意力聚合多视角时空信息，可直接复用于多视角动态场景的轨迹估计与几何一致性建模。
原文依据 MV-TAP utilizes camera geometry and a cross-view attention mechanism to aggregate spatio-temporal information across views
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论多视角相机系统能够对复杂真实场景提供丰富的观测，理解多视角设定下的动态目标已成为各类应用的核心。点跟踪是捕捉动态运动的关键机制。然而，传统单视角方法常因单目视频中可用的几何信息有限而失效，这成为向多视角扩展的关键瓶颈。本文提出 MV-TAP，一种鲁棒的点跟踪器，通过利用跨视角信息在多视角视频中跟踪动态场景的点。MV-TAP 利用相机几何与跨视角注意力机制聚合跨视角的时空信息，从而在多视角视频中实现更完整、更可靠的轨迹估计。为支撑该任务，我们构建了大规模合成训练数据集以及面向多视角跟踪的真实世界评测集。大量实验表明，MV-TAP 在具有挑战性的基准上优于现有点跟踪方法，为推进多视角点跟踪研究建立了有效的基线。
英文原文摘要 1147 字符
Multi-view camera systems enable rich observations of complex real-world scenes, and understanding dynamic objects in multi-view settings has become central to various applications. Point tracking serves as a key mechanism for capturing dynamic motion. However, conventional single-view approaches often fail due to the limited geometric information available in monocular video, which becomes a critical bottleneck for multi-view extension. In this work, we present MV-TAP, a robust point tracker that tracks points across multi-view videos of dynamic scenes by leveraging cross-view information. MV-TAP utilizes camera geometry and a cross-view attention mechanism to aggregate spatio-temporal information across views, enabling more complete and reliable trajectory estimation in multi-view videos. To support this task, we construct a large-scale synthetic training dataset and real-world evaluation sets tailored for multi-view tracking. Extensive experiments demonstrate that MV-TAP outperforms existing point-tracking methods on challenging benchmarks, establishing an effective baseline for advancing research in multi-view point tracking.