SmoCap: Movement Reconstruction under Morphology-Pose Ambiguity through Unified Scale-Pose Canonicalization

Abstract

Movement reconstruction pipelines need estimates that support interpretable joint motion and subject morphology, not only low marker fitting error. The same marker fitting error can be explained by different mixtures of morphology and posture, while weakly observed degrees of freedom are not uniquely identifiable, yielding anatomically inconsistent yet numerically acceptable solutions. We present SmoCap, a unified scale-pose framework that resolves morphology-posture ambiguity while preserving coordinated motion under weak observability. SmoCap solves a constrained trust-region QP with analytical proxy-mapped pose and scale Jacobians. Under matched observations, SmoCap is compared with an established OpenSim baseline using fluoroscopy-derived knee motion and anthropometric ground truth as external references. Extreme yoga sequences further probe coordinated spine motion under weak observability. In the controlled comparison, the OpenSim baseline achieved lower marker RMSE (8.81 vs 19.14 mm), whereas SmoCap achieved lower knee-orientation RMSE against fluoroscopy (5.82 deg vs 7.50 deg) in 24 of 28 trials and all six subjects. Mean absolute anthropometric endpoint errors were 8.44 vs 9.51 mm on CAMS-Knee and 6.87 vs 8.77 mm on Riglet for SmoCap and OpenSim, respectively. Proxy coupling preserved expressive and coordinated spine motion with marginal fitting error increase (+0.14 mm, +0.6%) in the yoga ablation. Median runtime was 0.204-0.332 ms/frame, with consistently 2-3 iterations. Marker RMSE alone did not reliably indicate better motion or morphology recovery. SmoCap instead combines unified scale-pose estimation with proxy coordination. It achieved lower motion and morphology errors in the external evaluations while supporting weak-observability-aware, cohort-scale movement reconstruction for downstream pipelines that depend on interpretable subject motion and morphology.