PAD-Hand: Physics-Aware Diffusion for Hand Motion Recovery

TL;DR

This paper introduces PAD-Hand, a physics-aware diffusion method that refines hand motion estimates from images into physically plausible sequences with interpretable confidence measures.

Contribution

It presents a novel physics-aware diffusion framework with a MeshCNN-Transformer backbone that models hand dynamics and estimates physics consistency variances.

Findings

Achieves consistent improvements over existing methods on hand datasets.

Provides interpretable variance maps indicating physical plausibility.

Produces physically plausible hand motions with confidence estimates.

Abstract

Significant advancements made in reconstructing hands from images have delivered accurate single-frame estimates, yet they often lack physics consistency and provide no notion of how confidently the motion satisfies physics. In this paper, we propose a novel physics-aware conditional diffusion framework that refines noisy pose sequences into physically plausible hand motion while estimating the physics variance in motion estimates. Building on a MeshCNN-Transformer backbone, we formulate Euler-Lagrange dynamics for articulated hands. Unlike prior works that enforce zero residuals, we treat the resulting dynamic residuals as virtual observables to more effectively integrate physics. Through a last-layer Laplace approximation, our method produces per-joint, per-time variances that measure physics consistency and offers interpretable variance maps indicating where physical consistency weakens. Experiments on two well-known hand datasets show consistent gains over strong image-based initializations and competitive video-based methods. Qualitative results confirm that our variance estimations are aligned with the physical plausibility of the motion in image-based estimates.