Objectivity without Ontological Commitment

Everyone seems to hate G.E. Moore's ethical intuitionism, because it assumes the existence of a non-natural property, 'good'. The real problem with doing this is that it makes the problematic assumption that we have some sort of bizarre epistemic access to this property, as did human beings during the creation of language itself, hence us making a word to describe this non-natural property.
On the other hand, it doesn't seem that non-cognitivists have our moral language right. When I am in an ethical discussion/argument with someone, it is more than just table pounding. Stevenson says ethical arguments are like preference arguments because we aren't stating physical facts to prove ethical facts ("We should go to the movie because it is an action movie"), but stating physical facts to attempt to convince the other person to hold our preferences ("[because I know you like action movies]"). But I'm not sure that an argument about anything, factual or relative, could go any other way.
At the end of the day, an ethical argument is just a matter of taste, say the non-cognitivists.

But that certainly doesn't seem right either.
So is there a way we can have objective moral truths without strange metaphysical commitments? YES.
It has been around for hundreds of years, mostly popular around Hume's time, but I am not sure why it failed to gain popularity when Roderick Firth formalized it in the 60s.
It is called the "Ideal Observer Theory". Under this theory, 'x is a good act' means 'the ideal observer would do x under these current conditions'. The ideal observer has certain characteristics, such as omniscience, impartiality, disinterest, etc. That is all important, but there is not enough space for that here. The point is, when we are making ethical arguments, we assume all these pretenses, such as utility, kindness, etc. The ideal observer has all of these characteristics, and puts its desires at the same level as others. But assuming these pretenses, an ethical dilemma does have an objective solution, so long as hypothetical situations can be considered true or false (for ex. If John likes pizza and is hungry, and a pizza is put in front of him, ceteris peribus he would eat the pizza).
One objection I will field briefly before I open this up for discussion is this: People can agree on all the facts of an ethical dilemma, and be unselfish, but still disagree on the outcome.
My response is this: Certainly it is possible, if not likely, that they have two different conceptions of the ideal observer. And these subtle differences could cause the difference. Not to mention, it is impossible for someone to know all facts about a situation, even if they were brilliant scientists--think about mental states and things like that. This opens up a lot of room for sensible disagreement. But when this happens between two people (most likely philosophers), all they have to do is step back and agree on a conception of an ideal observer, and then progress forward. This won't solve all problems, but it will help.

Also, I know nothing about aesthetics, but I'm not sure why the ideal observer theory wouldn't work in aesthetics. We have a piece of art: and we agree on a certain general conception of what good art (or music) is; takes creativity, emotional venting, etc. etc. But once we agree on a conception, we can ensure we are actually arguing about something instead of just talking past one another.