SPRig: Self-Supervised Pose-Invariant Rigging from Mesh Sequences

TL;DR

SPRig is a novel framework that enforces cross-frame consistency to learn pose-invariant rigging for dynamic mesh sequences, improving temporal coherence and reducing artifacts in skeleton and skinning generation.

Contribution

It introduces a general fine-tuning approach with novel regularization techniques for pose-invariant rigging on mesh sequences, applicable to both skeleton and skinning.

Findings

Achieves superior temporal coherence in rigging results.

Reduces artifacts compared to prior methods.

Enhances per-frame static quality without loss.

Abstract

State-of-the-art rigging methods typically assume a predefined canonical rest pose. However, this assumption does not hold for dynamic mesh sequences such as DyMesh or DT4D, where no canonical T-pose is available. When applied independently frame-by-frame, existing methods lack pose invariance and often yield temporally inconsistent topologies. To address this limitation, we propose SPRig, a general fine-tuning framework that enforces cross-frame consistency across a sequence to learn pose-invariant rigs on top of existing models, covering both skeleton and skinning generation. For skeleton generation, we introduce novel consistency regularization in both token space and geometry space. For skinning, we improve temporal stability through an articulation-invariant consistency loss combined with consistency distillation and structural regularization. Extensive experiments show that SPRig achieves superior temporal coherence and significantly reduces artifacts in prior methods, without sacrificing and often even enhancing per-frame static generation quality. The code is available in the supplemental material and will be made publicly available upon publication.