The authors propose CAST, a topology-aware motion capture framework from monocular video.
Motion capture from video works with any target skeleton topology
CAST combines learned rotation prediction with geometric correction to handle mismatched skeletons without retargeting.
Chinese Tech
Pu Pang · Zhongfei Qing · Ling-Hao Chen · Songnan Li · Hua Huang · Bo Hu
Xi'an Jiaotong University · Zhongguancun Academy · Tencent · Tsinghua University · Beijing Normal University
Research Digest··3 min read
The authors present CAST, a framework that recovers motion from a single monocular video onto a user-supplied target skeleton, even when that skeleton's topology—joint count, branching, and rest pose—differs from the subject in the video.
Why this paper
From Tencent and 4 others
In one line
CAST recovers motion from monocular video for any target skeleton topology using learned rotations and geometric correction.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ✓Limitations stated by the authors
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§