Chad Orzet has a great post on quantum mechanics and common sense. The position he takes is very Duhemian. That's perhaps not surprising; Duhem was a physicist himself (a thermodynamicist) who had thought long and hard about the relation between physics and common sense. Duhem's position on the subject can roughly be summarized in the following way:
(1) The practice of science, like all rational practice, is governed by common sense. This is the procedural point Orzet makes so well.
(2) Phyics is radically underdetermined by common-sense principles. Common sense is always consistent with lots of different possible physical theories because it is (generally) quite vague. But it's precisely because they're vague that common-sense principles can be obvious, right, and stable. Physical theory adds much-needed distinctness, but its precision comes at the price of being much less obvious, right, or stable.
(3) Bit by bit through the ages scientific practice adds to the depository of common sense. As scientific conclusions become more and more obviously right, however, they are assimilated to common sense. Even our common sense physics -- which is chiefly good for approximations in our general size-vicinity at practicable speeds -- is not read off of the world in any obvious way, but is the result of quite literally centuries of slow adaptation. (Arguably a reason why physics now seems so non-common-sensical, is that it has changed so quickly. Because common sense only includes things that have reached the point of clarity and obviousness to virtually all educated, rational persons, physics suddenly shot ahead of the slow pace at which common sense develops.) Common sense is in great measure a depository of science that has over the long years succeeded in becoming luminously clear, or, at least, that has succeeded in approaching such luminous clarity. What is more, on Duhem's view this is one of the twin goals of physics: the whole point of making a physical theory is (1) so it can be useful in the short term (for calculation, making new theories, etc.); and (2) so that, when your theory has in the long term begun to approach a state of perfection, it begins to be clearer and clearer how the theory relates the phenomena to the foundational principles of common sense. (This is also, according to Duhem, why we should be a bit suspicious of people claiming that a scientific conclusion is simply common sense; because more often than not, if they are right, it is because science has informed common sense rather than re-discovering something already there.)
(4) Common sense requires no justification. The principles of common sense are things that have become obvious. They are the very heart of what we call 'true and certain' -- the things that are foundationally, paradigmatically so. We aren't always right about what follows from these principles (in fact, we are often wrong), but the principles themselves are impeccable. Because of this, no skepticism can seriously defeat them. Duhem likes to quote and paraphrase Pascal on this: We have an impotence to prove that is invincible to any dogmatism and an idea of truth that is invincible to any skepticism. Common sense is the 'idea of truth' invincible to any skepticism, even though we cannot justify it in terms of anything more fundamental than itself. Common sense is what makes sense of things; a complete divorce between it and physics could never be possible, because such a divorce would mean that the 'physics' would be simply unintelligible, in every respect, to every rational person.
Some other things I've written on Duhem: Duhem's Realism; German Science; Teaching Physics by History.