MHR: Momentum Human Rig
🤖 AI Summary
Parameterized human models struggle to simultaneously achieve anatomical plausibility, expressive richness, and robust nonlinear pose correction within AR/VR and graphics pipelines. Method: We propose a novel model integrating ATLAS’s decoupled skeleton/shape representation with a momentum-inspired modern rigging architecture. It is the first to jointly incorporate anatomy-aware structural decoupling and data-driven nonlinear pose correction, enabling fine-grained deformation control under high degrees of freedom; combined with real-time skinning, it yields an end-to-end differentiable, lightweight, rendering-ready representation. Contribution/Results: Experiments demonstrate significant improvements in anatomical accuracy and visual realism for complex human poses, while maintaining computational efficiency and stability in real-time applications such as AR/VR. The model establishes a new paradigm for generative human modeling and interactive digital avatars.
Application Category
📝 Abstract
We present MHR, a parametric human body model that combines the decoupled skeleton/shape paradigm of ATLAS with a flexible, modern rig and pose corrective system inspired by the Momentum library. Our model enables expressive, anatomically plausible human animation, supporting non-linear pose correctives, and is designed for robust integration in AR/VR and graphics pipelines.
Problem
Research questions and friction points this paper is trying to address.
Develops parametric human body model for expressive animation
Enables anatomically plausible motion with pose correctives
Designed for robust AR/VR and graphics integration
Innovation
Methods, ideas, or system contributions that make the work stand out.
Parametric human model with decoupled skeleton/shape design
Flexible rig system incorporating Momentum pose correctives
Supports anatomically plausible animation for AR/VR pipelines
🔎 Similar Papers
No similar papers found.
