Automated game testing using computer vision methods

Video game development is a growing industry nowadays with high revenues. However, even if there are many resources invested in the software development process, many games still contain bugs or performance issues that affect the user experience. This paper presents ideas on how computer vision methods can be used to automate the process of game testing. The goal is to replace the parts of the testing process that require human users (testers) with machines as much as possible, in order to reduce costs and perform more tests in less time by scaling with hardware resources. The focus is on solving existing real-world problems that have emerged from several discussions with industry partners. We base our methods on previous work in this area using intelligent agents playing video games and deep learning methods that interpret feedback from their actions based on visual output. The paper proposes several methods and a set of open-source tools, independent of the operating system or deployment platform, to evaluate the efficiency of the presented methods.