A computational model for generating visually pleasing video game maps

Detalhes bibliográficos
Ano de defesa: 2016
Autor(a) principal: Hernandez Mariño, Julian Ricardo
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: eng
Instituição de defesa: Universidade Federal de Viçosa
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: http://www.locus.ufv.br/handle/123456789/8542
Resumo: In this work we introduce a computational model based on theories of graphical design to generate visually pleasing video game maps. We cast the problem of map generation as an optimization problem and prove it to be computationally hard. Then, we propose a heuristic search approach to solve the map generation problem and use it to generate levels of a clone of Super Mario Bros (SMB) called Infinite Mario Bros (IMB). Before evaluating the levels of IMB generated by our system, we perform a detailed study of the approaches commonly used to evaluate the content generated by computer programs. The evaluation used in previous works often relies on computational metrics. While these metrics are important for an initial exploratory evaluation of the content generated, it is not clear whether they are able to capture the player’s perception of the content generated. In this work we compare the insights gained from a user study with IMB levels generated by different systems with the insights gained from analyzing computational metric values. Our results suggest that current computational metrics should not be used in lieu of user studies for evaluating content generated by computer programs. Using the insights gained in our previous experiment, we performed another user study to evaluate the IMB levels generated by our method. The results show the advantage of our method over other approaches in terms of visual aesthetics and enjoyment. Finally, we performed one last user study that showed that our method is able to generate IMB levels with striking similarity to SMB levels created by professional designers.