Can smile-to-action liveness detection be replaced with different human interactions? (e.g. blink-to-action or nod-to-action) This functionality isn't currently supported.
