Geometry-Aware Humanoid Retargeting via Manifold ODEs
Geometry-Aware Humanoid Retargeting via Manifold ODEs
This paper introduces TaSP-KKT, a new method for interaction-preserving motion retargeting that transfers human demonstrations to humanoids while maintaining spatial relationships with objects and terrain. By modeling the continuous-time limit of frame-wise Karush-Kuhn-Tucker (KKT) conditions as an ordinary differential equation (ODE) on the robot configuration manifold, the approach connects existing temporally coupled constrained optimization methods to backward discretization of the ODE and proposes a forward discretization scheme. Numerical experiments indicate that TaSP-KKT achieves retargeting quality comparable to, though slightly lower than, OmniRetarget while running approximately twice as fast.
The paper proposes TaSP-KKT, a geometry-aware motion retargeting method for humanoids that models the continuous-time limit of frame-wise KKT conditions as an ODE on the robot configuration manifold, achieving comparable retargeting quality to OmniRetarget with approximately twice the speed by avoiding repeated SQP iterations.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
arXiv:2610.41349v1 Announce Type: new Abstract: In this paper, we focus on interaction-preserving motion retargeting, which transfers human demonstrations to humanoids while maintaining spatial relationships with objects and terrain. Previous work, such as OmniRetarget, addresses this task through temporally coupled constrained optimization, requiring repeated sequential quadratic programming (SQP) iterations at each
Open source S2
frame. We study this problem from an ordinary differential equation (ODE) perspective by modeling the continuous-time limit of its framewise Karush-Kuhn-Tucker (KKT) conditions as an ODE on the robot configuration manifold. This viewpoint offers two benefits: (i) it connects OmniRetarget's SQP iterations to a backward discretization of the ODE, enabling us to establish their local convergence under suitable regulari
Open source S3
ty assumptions; and (ii) it leads to TaSP-KKT (tangent-space parametrization for KKT flow), a new retargeting method based on a forward discretization of the KKT flow. The proposed method achieves the same first-order convergence to the ODE solution as the corresponding backward scheme while requiring only one weighted tangent-cone projection and a closed-form manifold update per step, thereby avoiding repeated inner
Open source S4
SQP iterations. Numerical experiments show that TaSP-KKT achieves retargeting quality comparable to, though slightly lower than, that of OmniRetarget while running approximately twice as fast.
Open source S5
Geometry-Aware Humanoid Retargeting via Manifold ODEs
Open source S1
Source：arXiv Robotics — research abstracts · arxiv.org