An Explorative Design Process for Game Map Generation Based on Satellite Images and Playability Factors

Georgi Nikolaev Ivanov, Magnus Håkon Petersen, Kristián Kovalsky, Kristian Mannerup Engberg Nielsen, George Palamas

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

Abstract

This paper proposes a tool for 3D Procedural Terrain Generation in games. The main objective of this study is to assess the quality of the generated terrains and thus the experience of the user. The tool measures how traversable a terrain is, how long it takes to walk across and how much of it is accessible. The produced results can be used to further refine the generation method. The evaluation consists of multiple terrains being produced, based on Generative Adversarial Networks (GANs) trained on satellite images. A path finding algorithm is then used to evaluate the traversability of each terrain between two points on the map. Results show that the proposed method can generate quite diverse and interesting maps and that the proposed metric can be used for evaluation of playability.
Original languageEnglish
Title of host publicationAn Explorative Design Process for Game Map Generation Based on Satellite Images and Playability Factors
Number of pages4
PublisherAssociation for Computing Machinery
Publication date15 Sep 2020
Pages1-4
Article number72
ISBN (Electronic)978-1-4503-8807-8
DOIs
Publication statusPublished - 15 Sep 2020
EventFDG '20: International Conference on the Foundations of Digital Games - Bugibba, Malta
Duration: 15 Sep 202018 Sep 2020
http://fdg2020.org/

Conference

ConferenceFDG '20: International Conference on the Foundations of Digital Games
CountryMalta
CityBugibba
Period15/09/202018/09/2020
Internet address

Keywords

  • generative adversarial networks
  • map generation
  • pathfinding
  • procedural generation

Fingerprint Dive into the research topics of 'An Explorative Design Process for Game Map Generation Based on Satellite Images and Playability Factors'. Together they form a unique fingerprint.

Cite this