In the field of camera motion research, it is widely held that the position (movement) and pose (rotation) of cameras are correlated and cannot be independently separated. A new equation based on inner product is proposed here to independently separate the position and pose. It is proved that the position and pose are not correlated and the equation is applied to estimation of the camera exterior parameters using a real image and 3D data.