That's a great point. They're certainly at least 3-dimensional, as seen most clearly when someone is on the main viewscreen but their eyeline matches the smaller figures on the bridge, rather than looking more like a zoom window as it would if it was a simple camera-flatscreen configuration.