AnthroTAP：用真实世界运动学习点跟踪
AnthroTAP: Learning Point Tracking with Real-World Motion
用 SMPL 网格顶点投影 + 光线投射遮挡判定 + 光流一致性过滤，从真实人体运动视频自动生成点跟踪伪标签，为视觉研究者提供了一条绕开昂贵人工标注、可扩展到其他可形变物体类别的真实数据监督构造思路。
原文依据 fits Skinned Multi-Person Linear (SMPL) models to detected humans, projects mesh vertices onto image planes, resolves occlusions via ray-casting
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论点跟踪模型往往难以泛化到真实世界视频，因为大规模训练数据以合成数据为主——这是目前唯一能够规模化生产的来源。然而，收集真实世界标注的成本高得难以承受，因为它需要在多帧之间跟踪数百个点。我们提出 AnthroTAP，一个自动化流水线，可从真实人体运动视频中生成大规模伪标注点跟踪数据。借助人体运动的结构化复杂性——非刚性形变、关节运动以及频繁遮挡——AnthroTAP 对检测到的人体拟合 Skinned Multi-Person Linear（SMPL）模型，将网格顶点投影到图像平面，通过光线投射解决遮挡，并利用光流一致性过滤不可靠的轨迹。在 AnthroTAP 数据集上训练的模型在 TAP-Vid 上取得最优性能，后者是一个具有挑战性的通用域基准，用于在多样化的刚性与非刚性物体（如人、动物、机器人和车辆）上跟踪任意点。我们的方法优于在规模大得多的真实数据集上训练的近期自训练方法，而仅需在 4 块 GPU 上训练一天。AnthroTAP 表明，结构化人体运动为点跟踪提供了一种可扩展且有效的真实世界监督来源。
英文原文摘要 1316 字符
Point tracking models often struggle to generalize to real-world videos because large-scale training data is predominantly synthetic--the only source currently feasible to produce at scale. Collecting real-world annotations, however, is prohibitively expensive, as it requires tracking hundreds of points across frames. We introduce AnthroTAP, an automated pipeline that generates large-scale pseudo-labeled point tracking data from real human motion videos. Leveraging the structured complexity of human movement--non-rigid deformations, articulated motion, and frequent occlusions--AnthroTAP fits Skinned Multi-Person Linear (SMPL) models to detected humans, projects mesh vertices onto image planes, resolves occlusions via ray-casting, and filters unreliable tracks using optical flow consistency. A model trained on the AnthroTAP dataset achieves state-of-the-art performance on TAP-Vid, a challenging general-domain benchmark for tracking any point on diverse rigid and non-rigid objects (e.g., humans, animals, robots, and vehicles). Our approach outperforms recent self-training methods trained on vastly larger real datasets, while requiring only one day of training on 4 GPUs. AnthroTAP shows that structured human motion offers a scalable and effective source of real-world supervision for point tracking.