2025/07/08 by Inès Hyeonsu Kim, Kim, Inès Hyeonsu, Seokju Cho +11 · 1 citation
Computer Science · Engineering · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Human Motion and Animation #Human Pose and Action Recognition #Video Surveillance and Tracking Methods
paper · pdf · doi:10.48550/arxiv.2507.06233
openalex publication_date 2025/07/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/02
Point tracking models often struggle to generalize to real-world videos because large-scale training data is predominantly synthetic\unicodex2014the 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\unicodex2014non-rigid deformations, articulated motion, and frequent occlusions\unicodex2014AnthroTAP 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.