CONTEXT-AWARE CONTENT GENERATION FOR VIRTUAL ENVIRONMENTS

Andrew Brock; Theodore Lim; James Millar Ritchie; Nick Weston

CONTEXT-AWARE CONTENT GENERATION FOR VIRTUAL ENVIRONMENTS

Andrew Brock
, Theodore Lim
, James Millar Ritchie
, Nick Weston

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Large scale scene generation is a computationally intensive operation, and added complexities arise when dynamic content generation is required. We propose a system capable of generating virtual content from non-expert input. The proposed system uses a 3-dimensional variational autoencoder to interactively generate new virtual objects by interpolating between extant objects in a learned low-dimensional space, as well as by randomly sampling in that space. We present an interface that allows a user to intuitively explore the latent manifold, taking advantage of the network’s ability to perform algebra in the latent space to help infer context and generalize to previously unseen inputs.

Original language	English
Title of host publication	Proceedings of the ASME 2016 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference 2016
Publisher	American Society of Mechanical Engineers
Pages	1-8
Number of pages	8
Publication status	Published - 21 Aug 2016
Event	36th ASME Computers and Information in Engineering Conference 2016 - North Carolina, Charlotte, United States Duration: 21 Aug 2016 → 24 Aug 2016 Conference number: 36th https://www.asme.org/events/idetccie

Conference

Conference	36th ASME Computers and Information in Engineering Conference 2016
Abbreviated title	CIE 2016
Country/Territory	United States
City	Charlotte
Period	21/08/16 → 24/08/16
Internet address	https://www.asme.org/events/idetccie

Keywords

Context-awareness
Deep Learning
Unsupervised learning
Convolutional neural networks (CNNs)
Virtual environments
Scene generation

Theodore Lim
- T.Limhw.acuk
- School of Engineering & Physical Sciences - Associate Professor
- School of Engineering & Physical Sciences, Institute of Mechanical, Process & Energy Engineering - Associate Professor
Person: Academic (Research & Teaching)

Cite this

@inproceedings{e282e3dc28e8448e87edb991d3d21f87,

title = "CONTEXT-AWARE CONTENT GENERATION FOR VIRTUAL ENVIRONMENTS",

abstract = "Large scale scene generation is a computationally intensive operation, and added complexities arise when dynamic content generation is required. We propose a system capable of generating virtual content from non-expert input. The proposed system uses a 3-dimensional variational autoencoder to interactively generate new virtual objects by interpolating between extant objects in a learned low-dimensional space, as well as by randomly sampling in that space. We present an interface that allows a user to intuitively explore the latent manifold, taking advantage of the network{\textquoteright}s ability to perform algebra in the latent space to help infer context and generalize to previously unseen inputs. ",

keywords = "Context-awareness, Deep Learning, Unsupervised learning, Convolutional neural networks (CNNs), Virtual environments, Scene generation",

author = "Andrew Brock and Theodore Lim and Ritchie, \{James Millar\} and Nick Weston",

year = "2016",

month = aug,

day = "21",

language = "English",

pages = "1--8",

booktitle = "Proceedings of the ASME 2016 International Design Engineering Technical Conferences \& Computers and Information in Engineering Conference 2016",

publisher = "American Society of Mechanical Engineers",

address = "United States",

note = "36th ASME Computers and Information in Engineering Conference 2016, CIE 2016 ; Conference date: 21-08-2016 Through 24-08-2016",

url = "https://www.asme.org/events/idetccie",

}

Brock, A, Lim, T, Ritchie, JM & Weston, N 2016, CONTEXT-AWARE CONTENT GENERATION FOR VIRTUAL ENVIRONMENTS. in Proceedings of the ASME 2016 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference 2016., IDETC2016-59997, American Society of Mechanical Engineers, pp. 1-8, 36th ASME Computers and Information in Engineering Conference 2016, Charlotte, United States, 21/08/16.

CONTEXT-AWARE CONTENT GENERATION FOR VIRTUAL ENVIRONMENTS. / Brock, Andrew; Lim, Theodore; Ritchie, James Millar et al.
Proceedings of the ASME 2016 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference 2016. American Society of Mechanical Engineers, 2016. p. 1-8 IDETC2016-59997.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - CONTEXT-AWARE CONTENT GENERATION FOR VIRTUAL ENVIRONMENTS

AU - Brock, Andrew

AU - Lim, Theodore

AU - Ritchie, James Millar

AU - Weston, Nick

N1 - Conference code: 36th

PY - 2016/8/21

Y1 - 2016/8/21

N2 - Large scale scene generation is a computationally intensive operation, and added complexities arise when dynamic content generation is required. We propose a system capable of generating virtual content from non-expert input. The proposed system uses a 3-dimensional variational autoencoder to interactively generate new virtual objects by interpolating between extant objects in a learned low-dimensional space, as well as by randomly sampling in that space. We present an interface that allows a user to intuitively explore the latent manifold, taking advantage of the network’s ability to perform algebra in the latent space to help infer context and generalize to previously unseen inputs.

AB - Large scale scene generation is a computationally intensive operation, and added complexities arise when dynamic content generation is required. We propose a system capable of generating virtual content from non-expert input. The proposed system uses a 3-dimensional variational autoencoder to interactively generate new virtual objects by interpolating between extant objects in a learned low-dimensional space, as well as by randomly sampling in that space. We present an interface that allows a user to intuitively explore the latent manifold, taking advantage of the network’s ability to perform algebra in the latent space to help infer context and generalize to previously unseen inputs.

KW - Context-awareness

KW - Deep Learning

KW - Unsupervised learning

KW - Convolutional neural networks (CNNs)

KW - Virtual environments

KW - Scene generation

M3 - Conference contribution

SP - 1

EP - 8

BT - Proceedings of the ASME 2016 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference 2016

PB - American Society of Mechanical Engineers

T2 - 36th ASME Computers and Information in Engineering Conference 2016

Y2 - 21 August 2016 through 24 August 2016

ER -

CONTEXT-AWARE CONTENT GENERATION FOR VIRTUAL ENVIRONMENTS

Abstract

Conference

Keywords

Fingerprint

Profiles

Theodore Lim

Cite this