Invertible Neural Skinning

TL;DR

This paper introduces Invertible Neural Skinning (INS), a novel method that enhances 3D human model reposing by learning pose-dependent deformations, preserving surface correspondences, and significantly improving speed and accuracy over existing techniques.

Contribution

The paper proposes a new invertible neural network architecture for skinning that extends linear blend skinning with learned pose-dependent deformations, enabling faster and more accurate reposing of clothed humans.

Findings

Outperforms state-of-the-art reposing methods

Preserves surface correspondences across poses

Runs an order of magnitude faster

Abstract

Building animatable and editable models of clothed humans from raw 3D scans and poses is a challenging problem. Existing reposing methods suffer from the limited expressiveness of Linear Blend Skinning (LBS), require costly mesh extraction to generate each new pose, and typically do not preserve surface correspondences across different poses. In this work, we introduce Invertible Neural Skinning (INS) to address these shortcomings. To maintain correspondences, we propose a Pose-conditioned Invertible Network (PIN) architecture, which extends the LBS process by learning additional pose-varying deformations. Next, we combine PIN with a differentiable LBS module to build an expressive and end-to-end Invertible Neural Skinning (INS) pipeline. We demonstrate the strong performance of our method by outperforming the state-of-the-art reposing techniques on clothed humans and preserving surface correspondences, while being an order of magnitude faster. We also perform an ablation study, which shows the usefulness of our pose-conditioning formulation, and our qualitative results display that INS can rectify artefacts introduced by LBS well. See our webpage for more details: https://yashkant.github.io/invertible-neural-skinning/