Learning Motion Manifolds with Convolutional Autoencoders

Daniel Holden, Jun Saito, Taku Komura, Thomas Joyce

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a technique for learning a manifold of human motion data using Convolutional Autoencoders. Our approach is capable of learning a manifold on the complete CMU database of human motion. This manifold can be treated as a prior probability distribution over human motion data, which has many applications in animation research, including projecting invalid or corrupt motion onto the manifold for removing error, computing similarity between motions using geodesic distance along the manifold, and interpolation of motion along the manifold for avoiding blending artefacts.
Original languageEnglish
Title of host publicationProceeding SA '15 SIGGRAPH Asia 2015 Technical Briefs
Place of PublicationNew York, NY, USA
PublisherACM
Number of pages4
ISBN (Print)978-1-4503-3930-8
DOIs
Publication statusPublished - 2015

Keywords

  • animation, autoencoding, character animation, convolutional neural networks, deep neural networks, machine learning, manifold learning, motion data

Fingerprint

Dive into the research topics of 'Learning Motion Manifolds with Convolutional Autoencoders'. Together they form a unique fingerprint.

Cite this