PIANO: A Parametric Hand Bone Model from Magnetic Resonance Imaging

TL;DR

PIANO is a novel, data-driven parametric model of human hand bones derived from MRI data, enabling more realistic biomechanics analysis and semantic understanding in VR, healthcare, and computer vision applications.

Contribution

It introduces the first biologically accurate, differentiable hand bone model from MRI data, enhancing anatomical precision over traditional outer-surface models.

Findings

Model is biologically correct and easy to animate.

Enables training neural networks with semantic loss for detailed hand bone understanding.

Open-sourced for community use.

Abstract

Hand modeling is critical for immersive VR/AR, action understanding, or human healthcare. Existing parametric models account only for hand shape, pose, or texture, without modeling the anatomical attributes like bone, which is essential for realistic hand biomechanics analysis. In this paper, we present PIANO, the first parametric bone model of human hands from MRI data. Our PIANO model is biologically correct, simple to animate, and differentiable, achieving more anatomically precise modeling of the inner hand kinematic structure in a data-driven manner than the traditional hand models based on the outer surface only. Furthermore, our PIANO model can be applied in neural network layers to enable training with a fine-grained semantic loss, which opens up the new task of data-driven fine-grained hand bone anatomic and semantic understanding from MRI or even RGB images. We make our model publicly available.