Evolving ambiguous images

Penousal Machado, Adriano Vinhas, João Correia, Aniko Ekárt

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

Abstract

This work explores the creation of ambiguous images, i.e., images that may induce multistable perception, by evolutionary means. Ambiguous images are created using a general purpose approach, composed of an expression-based evolutionary engine and a set of object detectors, which are trained in advance using Machine Learning techniques. Images are evolved using Genetic Programming and object detectors are used to classify them. The information gathered during classification is used to assign fitness. In a first stage, the system is used to evolve images that resemble a single object. In a second stage, the discovery of ambiguous images is promoted by combining pairs of object detectors. The analysis of the results highlights the ability of the system to evolve ambiguous images and the differences between computational and human ambiguous images.

Original languageEnglish
Title of host publicationProceedings of the twenty-fourth International Joint Conference on Artificial Intelligence (IJCAI 2015
PublisherInternational Joint Conferences on Artificial Intelligence
Pages2473-2479
Number of pages7
ISBN (Print)978-1-57735-738-4
Publication statusPublished - 2015
Event24th International Joint Conference on Artificial Intelligence - Buenos Aires, Argentina
Duration: 25 Jul 201531 Jul 2015

Conference

Conference24th International Joint Conference on Artificial Intelligence
Abbreviated titleIJCAI 2015
CountryArgentina
CityBuenos Aires
Period25/07/1531/07/15

Fingerprint

Detectors
Genetic programming
Learning systems
Engines

Bibliographical note

Funding: Fundação para a Ciêencia e Tecnologia (FCT), Portugal (SFRH/BD/90968/2012); and European Commission 7FP (project ConCreTe, Future and
Emerging Technologies (FET) (611733).

Cite this

Machado, P., Vinhas, A., Correia, J., & Ekárt, A. (2015). Evolving ambiguous images. In Proceedings of the twenty-fourth International Joint Conference on Artificial Intelligence (IJCAI 2015 (pp. 2473-2479). International Joint Conferences on Artificial Intelligence.
Machado, Penousal ; Vinhas, Adriano ; Correia, João ; Ekárt, Aniko. / Evolving ambiguous images. Proceedings of the twenty-fourth International Joint Conference on Artificial Intelligence (IJCAI 2015. International Joint Conferences on Artificial Intelligence, 2015. pp. 2473-2479
@inproceedings{797893edf31c4a9db8bcccdd4ce54c6c,
title = "Evolving ambiguous images",
abstract = "This work explores the creation of ambiguous images, i.e., images that may induce multistable perception, by evolutionary means. Ambiguous images are created using a general purpose approach, composed of an expression-based evolutionary engine and a set of object detectors, which are trained in advance using Machine Learning techniques. Images are evolved using Genetic Programming and object detectors are used to classify them. The information gathered during classification is used to assign fitness. In a first stage, the system is used to evolve images that resemble a single object. In a second stage, the discovery of ambiguous images is promoted by combining pairs of object detectors. The analysis of the results highlights the ability of the system to evolve ambiguous images and the differences between computational and human ambiguous images.",
author = "Penousal Machado and Adriano Vinhas and Jo{\~a}o Correia and Aniko Ek{\'a}rt",
note = "Funding: Funda{\cc}{\~a}o para a Ci{\^e}encia e Tecnologia (FCT), Portugal (SFRH/BD/90968/2012); and European Commission 7FP (project ConCreTe, Future and Emerging Technologies (FET) (611733).",
year = "2015",
language = "English",
isbn = "978-1-57735-738-4",
pages = "2473--2479",
booktitle = "Proceedings of the twenty-fourth International Joint Conference on Artificial Intelligence (IJCAI 2015",
publisher = "International Joint Conferences on Artificial Intelligence",

}

Machado, P, Vinhas, A, Correia, J & Ekárt, A 2015, Evolving ambiguous images. in Proceedings of the twenty-fourth International Joint Conference on Artificial Intelligence (IJCAI 2015. International Joint Conferences on Artificial Intelligence, pp. 2473-2479, 24th International Joint Conference on Artificial Intelligence, Buenos Aires, Argentina, 25/07/15.

Evolving ambiguous images. / Machado, Penousal; Vinhas, Adriano; Correia, João; Ekárt, Aniko.

Proceedings of the twenty-fourth International Joint Conference on Artificial Intelligence (IJCAI 2015. International Joint Conferences on Artificial Intelligence, 2015. p. 2473-2479.

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

TY - GEN

T1 - Evolving ambiguous images

AU - Machado, Penousal

AU - Vinhas, Adriano

AU - Correia, João

AU - Ekárt, Aniko

N1 - Funding: Fundação para a Ciêencia e Tecnologia (FCT), Portugal (SFRH/BD/90968/2012); and European Commission 7FP (project ConCreTe, Future and Emerging Technologies (FET) (611733).

PY - 2015

Y1 - 2015

N2 - This work explores the creation of ambiguous images, i.e., images that may induce multistable perception, by evolutionary means. Ambiguous images are created using a general purpose approach, composed of an expression-based evolutionary engine and a set of object detectors, which are trained in advance using Machine Learning techniques. Images are evolved using Genetic Programming and object detectors are used to classify them. The information gathered during classification is used to assign fitness. In a first stage, the system is used to evolve images that resemble a single object. In a second stage, the discovery of ambiguous images is promoted by combining pairs of object detectors. The analysis of the results highlights the ability of the system to evolve ambiguous images and the differences between computational and human ambiguous images.

AB - This work explores the creation of ambiguous images, i.e., images that may induce multistable perception, by evolutionary means. Ambiguous images are created using a general purpose approach, composed of an expression-based evolutionary engine and a set of object detectors, which are trained in advance using Machine Learning techniques. Images are evolved using Genetic Programming and object detectors are used to classify them. The information gathered during classification is used to assign fitness. In a first stage, the system is used to evolve images that resemble a single object. In a second stage, the discovery of ambiguous images is promoted by combining pairs of object detectors. The analysis of the results highlights the ability of the system to evolve ambiguous images and the differences between computational and human ambiguous images.

UR - http://www.scopus.com/inward/record.url?scp=84949815032&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84949815032

SN - 978-1-57735-738-4

SP - 2473

EP - 2479

BT - Proceedings of the twenty-fourth International Joint Conference on Artificial Intelligence (IJCAI 2015

PB - International Joint Conferences on Artificial Intelligence

ER -

Machado P, Vinhas A, Correia J, Ekárt A. Evolving ambiguous images. In Proceedings of the twenty-fourth International Joint Conference on Artificial Intelligence (IJCAI 2015. International Joint Conferences on Artificial Intelligence. 2015. p. 2473-2479