Video restoration is one, interesting, but rather specific, application area. What percentage of the world's overall computing efforts do you think will be devoted to that one area in, say, ten years from now? If your estimate is above 0.1% I think you're sadly mistaken.
The collection of things and ideas collectively covered by the vague term "cloud computing" (including "platform as a service", all kinds of "software as a service", "private clouds", and so forth) are a sea-change in computing comparable to those which, historically, marked the major shifts in computing's perspective -- from experimental machines devoted to a given researcher for a few hours at a time, to "batch processing", to "time-sharing", to personal computers (not to mention, as "side branches", minicomputers and the way the revolutionized laboratories and then factories, microprocessors embedded in growing number of devices, powerful mobile platforms known as "smart phones", ...).
What percentage of the world's overall computing efforts do you think will be using one or more of the approaches collectively described as "cloud computing" in, say, ten years from now? If your estimate is below 10% (a rather conservative "round number") I think you're laughably mistaken.
So, a ratio of 100:1 in terms of "overall interest" in a reasonable future timeframe is a very, very conservative estimate. How much your specific university work will impress prospective employers depends a lot, of course, on the quality of said work -- and it may be easier to make a substantial impact in a reasonably narrow area, than in a very broad one (just like Julius Caesar is alleged to have said "better be first in a Province, than second in Rome"!-). But in terms of overall interest and future job prospects in the two fields, per se, there is really no comparison!-)