← ArchivePaper2022

MoRig: Motion-Aware Rigging of Character Meshes from Point Clouds

Zhan Xu, Yang Zhou, Li Yi, Evangelos Kalogerakis

SIGGRAPH AsiaAcademic30 citesRigging

Rigs and animates 3D character meshes by encoding motion cues from single-view point cloud streams of a performing subject.

Open the paper ↗Trace in the lineage graph

Abstract

We present MoRig, a method that automatically rigs character meshes driven by single-view point cloud streams capturing the motion of performing characters. Our method is also able to animate the 3D meshes according to the captured point cloud motion. MoRig’s neural network encodes motion cues from the point clouds into features that are informative about the articulated parts of the performing character. These motion-aware features guide the inference of an appropriate skeletal rig for the input mesh, which is then animated based on the point cloud motion. Our method can rig and animate diverse characters, including humanoids, quadrupeds, and toys with varying articulation. It accounts for occluded regions in the point clouds and mismatches in the part proportions between the input mesh and captured character. Compared to other rigging approaches that ignore motion cues, MoRig produces more accurate rigs, well-suited for re-targeting motion from captured characters.

How to read this

Category: Method: neural rigging and animation from point clouds
Contributions: Encodes motion cues from single-view point cloud streams into motion-aware features informative about articulated parts
Uses those features to infer a skeletal rig for an input mesh and animate it from the captured motion
Handles diverse characters (humanoids, quadrupeds, toys) plus occlusions and proportion mismatches between mesh and captured subject
Context: Extends neural rigging in the vein of RigNet (Xu et al. 2020) by adding motion cues from point cloud streams rather than relying on the static mesh alone.Builds on: RigNet: Neural Rigging for Articulated Characters
Correctness: Validated qualitatively on a range of character types and against motion-agnostic rigging baselines, with explicit handling of occlusion and proportion mismatch; a reader should remember it is driven by single-view point clouds, so capture quality and viewpoint coverage bound the result.
Clarity: Accessible at a high level; a first pass conveys the motion-guided idea, a second pass is needed for the network design and rig inference.
How to read it: First pass to grasp how motion cues inform rig inference; second pass on the feature encoding and the rigging-then-animation pipeline if you plan to compare against RigNet-style static methods.

Builds on

RigNet: Neural Rigging for Articulated Characters 2020

Built upon by

Nothing yet.

Related work

RigNet: Neural Rigging for Articulated Characters 2020 / SIGGRAPH
S3: Neural Shape, Skeleton, and Skinning Fields for 3D Human Modeling 2021 / CVPR
Mobilizing Mocap, Motion Blending, and Mayhem: Rig Interoperability for Crowd Simulation on Incredibles 2 2018 / SIGGRAPH
One Model to Rig Them All: Diverse Skeleton Rigging with UniRig 2025 / SIGGRAPH

Keywords

character rigging motion-aware rigging point cloud sequence skeleton prediction skinning weights transformer motion encoder mesh deformation network correspondence module rignet linear blend skinning graph neural network gmedgenet pointnet++motion retargeting inverse kinematics articulated characters single-view performance capture

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 →