← ArchivePaper2007
Automatic Rigging and Animation of 3D Characters
Automatic skeleton embedding and skinning weight computation for arbitrary 3D meshes, enabling one-click rigging of novel characters.
Abstract
A system called Pinocchio automatically rigs 3D characters by embedding a skeleton inside the character geometry and computing bone weights using heat diffusion. The method discretizes skeleton embedding using an octree distance field and sphere packing, then refines placement with continuous optimization. Bone weights are computed by treating the character as a heat-conducting volume where bones maintain fixed temperatures, producing natural weight variation across the surface. The system handles diverse character proportions and has been tested on 16 unseen characters with high success rates.
How to read this
- Category
- Method / system: automatic skeleton rigging and skinning (Pinocchio)
- Contributions
- Automatic embedding of a skeleton inside arbitrary character geometry using an octree distance field, sphere packing, and continuous refinement.
- Bone-weight computation via heat diffusion, treating the character as a heat-conducting volume with bones at fixed temperatures.
- A working one-click rigging system tested across characters of diverse proportions.
- Context
- Sits in the lineage of automatic rigging and skinning-weight estimation, combining a geometric skeleton-embedding stage with a physically-motivated heat-diffusion weighting scheme.
- Correctness
- Demonstrated on a set of unseen characters with reported high success rates; the method assumes a reasonable match between the input mesh and the embedded skeleton template, so very atypical morphologies or topologies are a known caution.
- Clarity
- Clearly structured and accessible; a first pass conveys the two-stage approach, a second pass for the embedding optimization and the heat-diffusion weight formulation.
- How to read it
- First pass for the system overview (embed then weight); second pass on the discrete embedding and heat-equation weighting if you plan to reuse either stage independently.
Builds on
Nothing in the archive, this is a starting point.
Built upon by
- An Automated Pipeline for Generating Run-Time Rigs 2011
- Automatically Rigging Multi-Component Characters 2012
- Robust and Accurate Skeletal Rigging from Mesh Sequences 2014
- Avatar Reshaping and Automatic Rigging Using a Deformable Model 2015
- Animation Setup Transfer for 3D Characters 2016
- Segmentation-Based Skinning 2017
- KineFX for Games | Luiz Kruel | Houdini 18.5 HIVE 2020
- RigNet: Neural Rigging for Articulated Characters 2020
- SkinMixer: Blending 3D Animated Models 2022
Related work
- Geodesic Voxel Binding for Production Character Meshes 2013 / SCA
- Avatar Reshaping and Automatic Rigging Using a Deformable Model 2015 / MIG
- Mobilizing Mocap, Motion Blending, and Mayhem: Rig Interoperability for Crowd Simulation on Incredibles 2 2018 / SIGGRAPH
- Easy Rigging of Face by Automatic Registration and Transfer of Skinning Parameters 2010 / ICCVG
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 →