CAST: Real-Time Motion Capture for Any Skeleton Topology

cs.CV

Submitted: 2026-10-10

Updated: 2026-10-10

Comments: 39 pages. Code repository: https://github.com/fankewen/CAST

Code: https://github.com/fankewen/CAST

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• MoCapAnything V2: End-to-End Motion Capture for Arbitrary Skeletons
• HuMoT: Human Motion Representation using Topology-Agnostic Transformers for Character Animation Retargeting
• DINOv2: Learning Robust Visual Features without Supervision
• QuaterNet: A Quaternion-based Recurrent Model for Human Motion
• TopoCap: Learning Topology-Agnostic Motion Priors for Monocular Video-to-Animation
• DINOv3
• SAMoR: Motion Modelling for Articulated Objects of Any Skeleton and Topology

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models