← ArchivePaper2024
SKEL-Betweener: a Neural Motion Rig for Interactive Motion Authoring
Dhruv Agrawal, Jakob Buhmann, Dominik Borer, Robert W. Sumner, Martin Guay
Neural motion inbetweening that reasons about skeleton structure for intuitive, physics-plausible gap filling between sparse keyframes.
Abstract
Authoring 3D motions is a laborious process that requires manipulating and coordinating many control handles over time. Neural motion representations learned from large motion datasets have recently shown impressive capabilities in many motion completion tasks. However, current methods are not designed for interactive motion authoring workflows. The reasons being their requirement of a dense context of full poses, which takes considerable time to author, as well as their lack of joint-level controls for refinement. In this paper, we introduce a Neural Motion Rig called SKEL-Betweener, tailored to interactive motion authoring. SKEL-Betweener is able to generate long motion sequences from two poses only, and enables intermediate motion authoring via neural motion curves---intuitive joint-level controls for positions and orientations. Through user evaluations, we demonstrate the effectiveness of our Neural Motion Rig for efficiently creating and editing motions.
How to read this
- Category
- Method: neural motion rig for interactive in-betweening
- Contributions
- SKEL-Betweener, a neural motion rig tailored to interactive motion authoring
- Generation of long sequences from just two poses, removing the dense full-pose context other methods require
- Neural motion curves giving intuitive joint-level controls over positions and orientations for refinement
- Context
- An interactive-authoring take on learned in-betweening, building on Robust Motion In-Betweening (Harvey 2020) while adding skeleton-aware joint-level control.Builds on: Robust Motion In-Betweening
- Correctness
- Effectiveness shown through user evaluations of creating and editing motion; validation emphasizes authoring usability rather than large-scale quantitative motion benchmarks, so read the user study to judge the claims.
- Clarity
- Accessible and workflow-oriented; a first pass conveys the two-pose, joint-curve idea clearly.
- How to read it
- First pass for the authoring workflow (two poses plus neural motion curves) and the user-study findings; deeper pass only if you need the neural-rig representation internals.
Builds on
Built upon by
Related work
- Pose and Skeleton-aware Neural IK for Pose and Motion Editing 2023 / SIGGRAPH Asia
- Interactive Character Control with Auto-Regressive Motion Diffusion Models 2024 / TOG
- Factorized Motion Diffusion for Precise and Character-Agnostic Motion Inbetweening 2024 / MIG
- TEMOS: Generating Diverse Human Motions from Textual Descriptions 2022 / CVPR
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 →