I think the point is the assembling of random photos/videos into a convincing representation of textures and animation (without devoted hardware, bringing the person in for a scan, etc), not that 'this is bleeding edge visuals'.