← ArchivePaper2023

Pose and Skeleton-aware Neural IK for Pose and Motion Editing

Dhruv Agrawal, Martin Guay, Jakob Buhmann, Dominik Borer, Robert W. Sumner

SIGGRAPH AsiaDisney Research8 citesRigging ML Deformation

Neural IK solver aware of skeleton topology and pose context, enabling fast and accurate pose and motion editing for character rigs.

Open the paper ↗Trace in the lineage graph

Abstract

Posing a 3D character for film or game is an iterative and laborious process where many control handles (e.g. joints) need to be manipulated to achieve a compelling result. Neural Inverse Kinematics (IK) is a new type of IK that enables sparse control over a 3D character pose, and leverages full body correlations to complete the un-manipulated joints of the body. While neural IK is promising, current methods are not designed to preserve previous edits in posing workflows. Current models generate a single pose from the handles only, regardless of what was there previously, making it difficult to preserve any variations and hindering tasks such as pose and motion editing. In this paper, we introduce SKEL-IK, a novel architecture and training scheme that is conditioned on a base pose, and designed to flow information directly onto the skeletal graph structure, such that hard constraints can be enforced by blocking information flows at certain joints. As a result, we are able to satisfy both hard and soft constraints, as well as preserve un-manipulated parts of the body when desired. Finally, by controlling the base pose in different ways, we demonstrate the ability of our model to perform tasks such as generating variations and quickly editing poses and motions; with less erosion of the base poses compared to the current state-of-the-art.

How to read this

Category: Method: skeleton-aware neural inverse kinematics for pose/motion editing
Contributions: SKEL-IK, a neural IK architecture conditioned on a base pose so previous edits are preserved during posing workflows
Information flow routed onto the skeletal graph, allowing hard constraints by blocking flow at chosen joints while satisfying soft constraints
Preservation of un-manipulated body parts and support for pose and motion editing by controlling the base pose
Context: Builds on learned IK lineage such as Style-Based Inverse Kinematics (Grochow 2004), addressing prior neural-IK methods that ignore previous edits.Builds on: Style-Based Inverse Kinematics
Correctness: Targets the realistic workflow limitation that neural IK overwrites prior edits; benefits depend on the base-pose conditioning and graph-blocking design, so generalization across rigs should be confirmed.
Clarity: Moderately technical; a first pass conveys the base-pose-conditioned, graph-structured idea, a second pass clarifies the constraint-enforcement mechanics.
How to read it: First pass for the edit-preservation problem and the skeletal-graph blocking idea; second pass on the architecture and constraint handling if you build posing or IK tools.

Builds on

Style-Based Inverse Kinematics 2004

Built upon by

Nothing yet.

Related work

SketchiMo: Sketch-based Motion Editing for Articulated Characters 2016 / SIGGRAPH
SKEL-Betweener: a Neural Motion Rig for Interactive Motion Authoring 2024 / SIGGRAPH Asia
How the Rig Design Impacts the Animation Process 2021 / DigiPro
How to Train Your Dog: Neural Enhancement of Quadruped Animations 2021 / MIG

Keywords

inverse kinematics neural network pose preservation skeleton-aware graph transformer skeletal graph pose authoring motion editing constraint satisfaction graph neural network multi-resolution hard constraints soft constraints forward kinematics pose completion

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 →