Affect-aware procedural game generation
My proposed research lies at the intersection of affective computing and machine learning with the intelligent design of games, and in particular, procedurally generated and player- adaptive game content. In essence, the main aim of this research would be to develop suitable techniques for assessing the player's emotional state by utilising various sources of information, such as e.g., the facial expressions, gestures and non-verbal cues of the player along with the actual game footage (pixel values). By utilising the model of the player’s behaviour and reactions, the generation of content will be tuned thus providing a personalised experience to the player, while also being able to provide useful knowledge to the developers, e.g. by predicting the expected reaction of a player given game footage. The potential impact of this research spans several practical applications, from personalised gameplay by means of player-adaptive games to providing new tools for procedural game-content generation based on generative machine learning.
Born and raised on the sunny Isle of Wight (a small island off the south coast of England) I recently graduated from Goldsmiths with a BSc Computer Science. Prior coming to Goldsmiths I worked for several years at Stainless Games first as in Quality Assurance and then later as a Junior Programmer, mostly on the Magic: The Gathering series. I had never written a line of code before working as a programmer but I fell in love with game development, programming and AI. I use dark themes whenever possible.
Home institution: Goldsmiths
Supervisor: Dr Mihalis Nicolaou
Ready to apply?
Once you have identified your potential supervisor, we would encourage you to contact them to discuss your research proposal.Learn More