Can we tell something about a person from the way they move, even when their appearance has been completely changed?
In this project, the student will use selected interactions from Meta’s Seamless Interaction dataset to create animated MetaHuman characters in Unreal Engine 5. Real conversational motion will be transferred onto virtual characters whose appearance differs from that of the original performer, for example by changing gender presentation or apparent age.
The student will create a controlled set of video stimuli in which the same underlying motion is shown on different MetaHuman identities. A perceptual experiment will then investigate whether observers can detect characteristics of the original performer from motion alone, or whether the appearance of the virtual character dominates their judgement.
Possible questions include whether participants can match animations produced by the same performer, estimate characteristics of the underlying performer, or distinguish between motion captured from different demographic groups when visual appearance is held constant or deliberately mismatched.
The project combines Unreal Engine 5, MetaHumans, character animation and human perception, and would suit a student interested in computer graphics, animation or experimental studies of virtual humans.