Automatic Similarity Detection in LEGO Ducks

Mark Ferguson; Christoph Sebastian Deterding; Andreas Lieberoth; Marc Malmdorf Andersen; Sam Devlin; Daniel Kudenko; James Alfred Walker

Automatic Similarity Detection in LEGO Ducks

Mark Ferguson, Christoph Sebastian Deterding, Andreas Lieberoth, Marc Malmdorf Andersen, Sam Devlin, Daniel Kudenko, James Alfred Walker

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

Abstract

The automated evaluation of creative products promises both good-and-scalable creativity assessments and new forms of visual analysis of whole corpora. Where creative works are not ‘born digital’, such automated evaluation requires fast and frugal ways of transforming them into data representations that can be meaningfully assessed with common creativity metrics like novelty. In this paper, we report the results of training a Spatiotemporal DeepInfomax Variational Autoencoder (STDIM-VAE) on a digital photo pool of 162 LEGO ducks to generate a phenotypical landscape of clusters of similar ducks and dissimilarity scores for individual ducks. Visual inspection suggests that our system produces plausible results from image pixels alone. We conclude that under certain conditions, STDIM-VAEs may provide fast and frugal ways of automatically assessing corpora of creative works.

Original language	English
Title of host publication	ICCC'20: Eleventh International Conference on Computational Creativity
Publisher	Association for Computational Creativity (ACC)
Number of pages	4
Publication status	Accepted/In press - 30 Jun 2020

Access to Document

Automatic_Similarity_Detection_LEGO_DucksAccepted author manuscript, 497 KBLicence: Other

1 Finished
1 Active

EPSRC IGGI CDT
Cairns, P. A. & Cowling, P. I.
EPSRC
1/04/14 → 31/07/25
Project: Research project (funded) › Studentship (central)
The Digital Creativity Hub
Cowling, P. I., Austin, J., Cairns, P. A., Holliman, N. S., Hook, J. D., Marsden, E., Murphy, D. T., Petrie, H., Reed, D. J., Richards, J. D., Ursu, M., Wade, A., Baier, H., Beale, G., Block, F. O., Deterding, C. S., Devlin, S., Drachen, A., Kasprowicz, R. E., Smith, D. P. & Williams, D. A. H.
EPSRC
1/10/15 → 30/09/22
Project: Research project (funded) › Research

Cite this

@inproceedings{26369ba6dd6b442ebeae14d5f501bb7a,

title = "Automatic Similarity Detection in LEGO Ducks",

abstract = "The automated evaluation of creative products promises both good-and-scalable creativity assessments and new forms of visual analysis of whole corpora. Where creative works are not {\textquoteleft}born digital{\textquoteright}, such automated evaluation requires fast and frugal ways of transforming them into data representations that can be meaningfully assessed with common creativity metrics like novelty. In this paper, we report the results of training a Spatiotemporal DeepInfomax Variational Autoencoder (STDIM-VAE) on a digital photo pool of 162 LEGO ducks to generate a phenotypical landscape of clusters of similar ducks and dissimilarity scores for individual ducks. Visual inspection suggests that our system produces plausible results from image pixels alone. We conclude that under certain conditions, STDIM-VAEs may provide fast and frugal ways of automatically assessing corpora of creative works.",

author = "Mark Ferguson and Deterding, {Christoph Sebastian} and Andreas Lieberoth and {Malmdorf Andersen}, Marc and Sam Devlin and Daniel Kudenko and Walker, {James Alfred}",

year = "2020",

month = jun,

day = "30",

language = "English",

booktitle = "ICCC'20: Eleventh International Conference on Computational Creativity",

publisher = "Association for Computational Creativity (ACC)",

}

TY - GEN

T1 - Automatic Similarity Detection in LEGO Ducks

AU - Ferguson, Mark

AU - Deterding, Christoph Sebastian

AU - Lieberoth, Andreas

AU - Malmdorf Andersen, Marc

AU - Devlin, Sam

AU - Kudenko, Daniel

AU - Walker, James Alfred

PY - 2020/6/30

Y1 - 2020/6/30

N2 - The automated evaluation of creative products promises both good-and-scalable creativity assessments and new forms of visual analysis of whole corpora. Where creative works are not ‘born digital’, such automated evaluation requires fast and frugal ways of transforming them into data representations that can be meaningfully assessed with common creativity metrics like novelty. In this paper, we report the results of training a Spatiotemporal DeepInfomax Variational Autoencoder (STDIM-VAE) on a digital photo pool of 162 LEGO ducks to generate a phenotypical landscape of clusters of similar ducks and dissimilarity scores for individual ducks. Visual inspection suggests that our system produces plausible results from image pixels alone. We conclude that under certain conditions, STDIM-VAEs may provide fast and frugal ways of automatically assessing corpora of creative works.

AB - The automated evaluation of creative products promises both good-and-scalable creativity assessments and new forms of visual analysis of whole corpora. Where creative works are not ‘born digital’, such automated evaluation requires fast and frugal ways of transforming them into data representations that can be meaningfully assessed with common creativity metrics like novelty. In this paper, we report the results of training a Spatiotemporal DeepInfomax Variational Autoencoder (STDIM-VAE) on a digital photo pool of 162 LEGO ducks to generate a phenotypical landscape of clusters of similar ducks and dissimilarity scores for individual ducks. Visual inspection suggests that our system produces plausible results from image pixels alone. We conclude that under certain conditions, STDIM-VAEs may provide fast and frugal ways of automatically assessing corpora of creative works.

M3 - Conference contribution

BT - ICCC'20: Eleventh International Conference on Computational Creativity

PB - Association for Computational Creativity (ACC)

ER -

Automatic Similarity Detection in LEGO Ducks

Abstract

Access to Document

Projects

EPSRC IGGI CDT

The Digital Creativity Hub

Cite this