Automated Aesthetic Selection of Evolutionary Art by
Distance Based Classification of Genomes and Phenomes
using the Universal Similarity Metric
N. Svangard, and P. Nordin. Applications of Evolutionary Computing,
EvoWorkshops2004: EvoBIO, EvoCOMNET, EvoHOT,
EvoIASP, EvoMUSART, EvoSTOC, volume 3005 of LNCS, page 447--456. Coimbra, Portugal, Springer Verlag, (5-7 April 2004)
Abstract
We present a technique for automatically approximating
the aesthetic fitness of evolutionary art. Instead of
assigning fitness values to images interactively, we
use the Universal Similarity Metric to predict how
interesting new images are to the observer based on a
library of aesthetic images. In order to approximate
the Information Distance, and find the images most
similar to the training set, we use a combination of
zip-compression, for genomes, and jpeg-compression of
the final images. We evaluated the prediction accuracy
of our system by letting the user label a new set of
images and then compare that to what our system
classifies as the most aesthetically pleasing images.
Our experiments indicate that the Universal Similarity
Metric can successfully be used to classify what images
and genomes are aesthetically pleasing, and that it can
clearly distinguish between ügly" and "pretty"
images with an accuracy better than the random
baseline.
%0 Conference Paper
%1 svangard:evows04
%A Svangard, Nils
%A Nordin, Peter
%B Applications of Evolutionary Computing,
EvoWorkshops2004: EvoBIO, EvoCOMNET, EvoHOT,
EvoIASP, EvoMUSART, EvoSTOC
%C Coimbra, Portugal
%D 2004
%E Raidl, Guenther R.
%E Cagnoni, Stefano
%E Branke, Jurgen
%E Corne, David W.
%E Drechsler, Rolf
%E Jin, Yaochu
%E Johnson, Colin R.
%E Machado, Penousal
%E Marchiori, Elena
%E Rothlauf, Franz
%E Smith, George D.
%E Squillero, Giovanni
%I Springer Verlag
%K algorithms, computation evolutionary genetic programming,
%P 447--456
%T Automated Aesthetic Selection of Evolutionary Art by
Distance Based Classification of Genomes and Phenomes
using the Universal Similarity Metric
%U http://www.springerlink.com/openurl.asp?genre=article&issn=0302-9743&volume=3005&spage=447
%V 3005
%X We present a technique for automatically approximating
the aesthetic fitness of evolutionary art. Instead of
assigning fitness values to images interactively, we
use the Universal Similarity Metric to predict how
interesting new images are to the observer based on a
library of aesthetic images. In order to approximate
the Information Distance, and find the images most
similar to the training set, we use a combination of
zip-compression, for genomes, and jpeg-compression of
the final images. We evaluated the prediction accuracy
of our system by letting the user label a new set of
images and then compare that to what our system
classifies as the most aesthetically pleasing images.
Our experiments indicate that the Universal Similarity
Metric can successfully be used to classify what images
and genomes are aesthetically pleasing, and that it can
clearly distinguish between ügly" and "pretty"
images with an accuracy better than the random
baseline.
%@ 3-540-21378-3
@inproceedings{svangard:evows04,
abstract = {We present a technique for automatically approximating
the aesthetic fitness of evolutionary art. Instead of
assigning fitness values to images interactively, we
use the Universal Similarity Metric to predict how
interesting new images are to the observer based on a
library of aesthetic images. In order to approximate
the Information Distance, and find the images most
similar to the training set, we use a combination of
zip-compression, for genomes, and jpeg-compression of
the final images. We evaluated the prediction accuracy
of our system by letting the user label a new set of
images and then compare that to what our system
classifies as the most aesthetically pleasing images.
Our experiments indicate that the Universal Similarity
Metric can successfully be used to classify what images
and genomes are aesthetically pleasing, and that it can
clearly distinguish between {"}ugly{"} and {"}pretty{"}
images with an accuracy better than the random
baseline.},
added-at = {2008-06-19T17:46:40.000+0200},
address = {Coimbra, Portugal},
author = {Svangard, Nils and Nordin, Peter},
biburl = {https://www.bibsonomy.org/bibtex/27b2c66a6dc599bae4e2db975c14b1b3c/brazovayeye},
booktitle = {Applications of Evolutionary Computing,
EvoWorkshops2004: {EvoBIO}, {EvoCOMNET}, {EvoHOT},
{EvoIASP}, {EvoMUSART}, {EvoSTOC}},
editor = {Raidl, Guenther R. and Cagnoni, Stefano and Branke, Jurgen and Corne, David W. and Drechsler, Rolf and Jin, Yaochu and Johnson, Colin R. and Machado, Penousal and Marchiori, Elena and Rothlauf, Franz and Smith, George D. and Squillero, Giovanni},
interhash = {f14cfe3861ae295368c122d6b4b6908d},
intrahash = {7b2c66a6dc599bae4e2db975c14b1b3c},
isbn = {3-540-21378-3},
keywords = {algorithms, computation evolutionary genetic programming,},
month = {5-7 April},
notes = {EvoWorkshops2004},
pages = {447--456},
publisher = {Springer Verlag},
publisher_address = {Berlin},
series = {LNCS},
timestamp = {2008-06-19T17:52:29.000+0200},
title = {Automated Aesthetic Selection of Evolutionary Art by
Distance Based Classification of Genomes and Phenomes
using the Universal Similarity Metric},
url = {http://www.springerlink.com/openurl.asp?genre=article&issn=0302-9743&volume=3005&spage=447},
volume = 3005,
year = 2004
}