The kind of stabilisation in those lenses (though I'm not an expert in that) strikes me as being wholly unsuitable to the kind of stabilisation you want.
A ship moves in three axes - roll, pitch and yaw, while it also heaves, drifts and surges and hogs/sags and will occasionally alter course (heading). Roll is the most pronounced periodic movement, e.g. a 5000T ship rolls (from memory) with a period of ~10 secs through, say, +/- ~5 deg in typical seas though hull design and stabilizers have an effect on amplitude and period/pattern of roll. Larger ships will typically have a slower roll period and smaller amplitude. On a large, stable cruise ship, you may get away with only having to worry about roll, sea-state permitting.
Image stabilisation in a camera lens or body ("anti-shake") operates at a much higher frequency (I think it's typically in the 10Hz range) and smaller amplitude and I doubt that ship movement will even register on a camera system given a frequency factor of the order of 1/100x.
Any accurate pointing system on a ship requires gyro-stabilisation or accelerometer-stabilisation, either being mounted on a stabilised platform or with two-axis drive using differential pointing inputs (commanded direction in bearing and elevation, or alt/az if you prefer,and gyro pitch/roll/heading inputs). (Just for completeness ... electronic beam steering via phased-arrays is also used, but that still needs a gyro input and operates just like a mechanical system in that respect).
Without the above kind of system, I don't know how you'd manage the pointing variation in your images. On particularly calm days, with roll <<1 deg, you might get some useful shots. Possibly you could rig a passive system (hammock style) in one axis only, but it would be imperfect and stabilizers would throw it off a bit, as they can be act out of sync with the near-sinusoidal roll. Plus the ocean tends to have a mind of its own, sometimes.
I suppose manual selection of images from hundreds or thousands would be too time consuming. If someone else has managed to do at-sea timelapse without a stabilised plarform, then I can only think they had especially calm days on a large, stable vessel or that they hand-selected their images, or maybe had some very clever software that did the hard work for them.
I can't really offer you a solution, but perhaps this puts some bounds around the problem.