Skip to content

← ArchivePaper2023

Robust Skin Weights Transfer via Weight Inpainting

Rinat Abdrashitov, Kim Raichstat, Jared Monsen, David Hill

SIGGRAPH AsiaEpic13 citesSkinningRigging

Fully automatic two-stage method for transferring skin weights between geometrically dissimilar meshes, outperforming commercial software on garments.

Abstract

We present a new method for the robust transfer of skin weights from a source mesh to a target mesh with significantly different geometric shapes. Rigging garments is a typical application of skin weight transfer where weights are copied from a source body mesh to avoid tedious weight painting from scratch. However, existing techniques struggle with non-skin-tight garments and require additional manual weight painting. We introduce a fully automatic two-stage skin weight transfer process. First, an initial transfer is performed by copying weights from the source mesh only for those vertices on the target mesh where we have high confidence in obtaining the ground truth weights from the source. Then, we automatically compute weights for all other vertices by interpolating the weights computed in stage one. This approach is robust and easy to implement in practice, yet it far outperforms the methods used in existing commercial software and previous research works.

How to read this

Category
Method: an automatic skin-weight transfer algorithm
Contributions
  • A fully automatic two-stage skin weight transfer from a source mesh to a geometrically dissimilar target
  • Stage one copies weights only for high-confidence target vertices, stage two interpolates weights for the rest (weight inpainting)
  • Robustness on non-skin-tight garments, reported to outperform existing commercial software and prior research
Context
Addresses production garment rigging in the lineage of weight-transfer and binding methods such as Geodesic Voxel Binding (Dionne et al.).Builds on: Geodesic Voxel Binding for Production Character Meshes
Correctness
The key assumption is that a confident subset of correspondences gives reliable ground-truth weights from which the remainder can be interpolated; demonstrated mainly on garments over body meshes, so very loose or topologically very different targets are worth checking.
Clarity
Accessible and described as easy to implement; a first pass conveys the two-stage idea, a second pass clarifies the confidence criterion and inpainting.
How to read it
Focus on how high-confidence vertices are selected and how interpolation fills the rest; one careful pass is usually enough to apply it, a second pass for exact implementation details.

Builds on

Built upon by

Nothing yet.

Related work

Keywords

This page summarises the entry and links to its original source. The archive never hosts or redistributes the publication itself.Show it in the full archive list →