End-to-End Motion Capture from Rigid Body Markers with Geodesic Loss

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Lan, Hai, Li, Zongyan, Hu, Jianmin, Yang, Jialing, Dai, Houde
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917096059830272
author Lan, Hai
Li, Zongyan
Hu, Jianmin
Yang, Jialing
Dai, Houde
author_facet Lan, Hai
Li, Zongyan
Hu, Jianmin
Yang, Jialing
Dai, Houde
contents Marker-based optical motion capture (MoCap), while long regarded as the gold standard for accuracy, faces practical challenges, such as time-consuming preparation and marker identification ambiguity, due to its reliance on dense marker configurations, which fundamentally limit its scalability. To address this, we introduce a novel fundamental unit for MoCap, the Rigid Body Marker (RBM), which provides unambiguous 6-DoF data and drastically simplifies setup. Leveraging this new data modality, we develop a deep-learning-based regression model that directly estimates SMPL parameters under a geodesic loss. This end-to-end approach matches the performance of optimization-based methods while requiring over an order of magnitude less computation. Trained on synthesized data from the AMASS dataset, our end-to-end model achieves state-of-the-art accuracy in body pose estimation. Real-world data captured using a Vicon optical tracking system further demonstrates the practical viability of our approach. Overall, the results show that combining sparse 6-DoF RBM with a manifold-aware geodesic loss yields a practical and high-fidelity solution for real-time MoCap in graphics, virtual reality, and biomechanics.
format Preprint
id arxiv_https___arxiv_org_abs_2511_16418
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle End-to-End Motion Capture from Rigid Body Markers with Geodesic Loss
Lan, Hai
Li, Zongyan
Hu, Jianmin
Yang, Jialing
Dai, Houde
Computer Vision and Pattern Recognition
Human-Computer Interaction
Marker-based optical motion capture (MoCap), while long regarded as the gold standard for accuracy, faces practical challenges, such as time-consuming preparation and marker identification ambiguity, due to its reliance on dense marker configurations, which fundamentally limit its scalability. To address this, we introduce a novel fundamental unit for MoCap, the Rigid Body Marker (RBM), which provides unambiguous 6-DoF data and drastically simplifies setup. Leveraging this new data modality, we develop a deep-learning-based regression model that directly estimates SMPL parameters under a geodesic loss. This end-to-end approach matches the performance of optimization-based methods while requiring over an order of magnitude less computation. Trained on synthesized data from the AMASS dataset, our end-to-end model achieves state-of-the-art accuracy in body pose estimation. Real-world data captured using a Vicon optical tracking system further demonstrates the practical viability of our approach. Overall, the results show that combining sparse 6-DoF RBM with a manifold-aware geodesic loss yields a practical and high-fidelity solution for real-time MoCap in graphics, virtual reality, and biomechanics.
title End-to-End Motion Capture from Rigid Body Markers with Geodesic Loss
topic Computer Vision and Pattern Recognition
Human-Computer Interaction
url https://arxiv.org/abs/2511.16418