← ArchivePaper2016
Animation Setup Transfer for 3D Characters
Quentin Avril, Donya Ghafourzadeh, Srinivasan Ramachandran, Sahel Fallahdoust, Sarah Ribet, Olivier Dionne, Martin de Lasa, Eric Paquette
Transfers skeleton structure and skinning weights between characters with different mesh topologies without manual re-rigging.
Abstract
We present a general method for transferring skeletons and skinning weights between characters with distinct mesh topologies. Our pipeline takes as inputs a source character rig (consisting of a mesh, a transformation hierarchy of joints, and skinning weights) and a target character mesh. From these inputs, we compute joint locations and orientations that embed the source skeleton in the target mesh, as well as skinning weights to bind the target geometry to the new skeleton. Our method consists of two key steps. We first compute the geometric correspondence between source and target meshes using a semi‐automatic method relying on a set of markers. The resulting geometric correspondence is then used to formulate attribute transfer as an energy minimization and filtering problem. We demonstrate our approach on a variety of source and target bipedal characters, varying in mesh topology and morphology. Several examples demonstrate that the target characters behave well when animated with either forward or inverse kinematics. Via these examples, we show that our method preserves subtle artistic variations; spatial relationships between geometry and joints, as well as skinning weight details, are accurately maintained. Our proposed pipeline opens up many exciting possibilities to quickly animate novel characters by reusing existing production assets.
How to read this
- Category
- Method: skeleton and skinning-weight transfer between characters
- Contributions
- Transfers a source rig (mesh, joint hierarchy, skinning weights) onto a target mesh with distinct topology, computing embedded joint locations/orientations and new skinning weights
- Establishes source-to-target geometric correspondence via a semi-automatic marker-based method
- Formulates attribute transfer as an energy-minimization and filtering problem, validated on varied bipedal characters under both forward and inverse kinematics
- Context
- Continues the automatic rigging lineage of Baran and Popovic's Pinocchio (baran-automatic-rigging-2007), focusing on reusing an existing rig rather than building one from scratch.Builds on: Automatic Rigging and Animation of 3D Characters
- Correctness
- Relies on a few manually placed markers to seed correspondence and is demonstrated on bipedal characters varying in topology and morphology; a reader should note results are shown for bipeds and that quality depends on marker placement and source-target morphological similarity.
- Clarity
- Accessible; a first pass conveys the two-step (correspondence then transfer) pipeline, a second pass covers the energy formulation.
- How to read it
- First pass for the correspondence-then-transfer structure and what inputs it needs; second pass on the energy-minimization step if you want to reproduce the weight and joint transfer.
Builds on
Built upon by
Nothing yet.
Related work
- SkinMixer: Blending 3D Animated Models 2022 / SIGGRAPH Asia
- Sculpt Processing for Character Rigging 2020 / SIGGRAPH
- A Statistical Model of Human Pose and Body Shape 2009 / CGF
- Automatic Rigging and Animation of 3D Characters 2007 / SIGGRAPH
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 →