The eyes are the window to the soul, so they say.
A massive part of our brain is dedicated to the visual analysis of our world, which is probably why we feel that the centre of our being is in our head – I think this feeling is more related to the location of our most important sense than to the actual location of our brain. A hippy friend once said to me after he'd bumped his head on a short doorframe, "I always forget I have a body above my eyes". he was stoned at the time though.
The reason we need a lot of brainspace for our vision is that it is a very complex mathematical problem to solve.
If you have only one eye then it is presented with a 2D image with no depth information – your depth perception is certainly much reduced with only one eye, but you can navigate your way around day to day life without difficulty.
This is something that is not to be sniffed at – your brain has to make many many assumptions about what an object's shape actually is and some shading information to then turn this into a sensible image presented to your conscious mind. For example, you know a TV screen is a rectangle, but if you see it from an angle it will form an image of a parallelogram onto your retina. It seems reasonable to assume it should actually be a rectangle and therefore is interpreted as a rectangle at an angle. What if it isn't an object you've seen before... a lot of processing is needed and optical illusions show some of the limitations of that processing.
When you have 2 eyes depth perception becomes easier, because each eye sees a slightly different image due to parallax. However, turning these 2 images into a coherent 3D world is nothing to be sneezed at either, and even the very best computer systems with artificial vision (2 camera 'eyes') are very slow and have a high failure rate compared to a human being.
The computational theory of mind is still the best theory to explain how we do things, and the current limitations on artificial vision are not evidence against it, but merely highlight what a massively complex calculation it is, and that it does require some reasonable assumptions – vision is an incomplete information problem.
One interesting validation of the computational theory of mind to do with vision that I read about recently concerned the reaction times of recognising shapes. Someone is asked to memorise a list of a few strange symbols, and then shapes are flashed onto a screen. When one of the ones in the list appears the subject presses a button. If the shapes are rotated when they are flashed up then the time it takes to recognise it is linearly proportional to the angle it was rotated by. This seems reasonable if the brain has a 'module' that rotates shapes. If the shape is a mirror image and rotated then the reaction time is exactly the same no matter what the rotation. At first this puzzled the researchers but a mathematician pointed out to them that for a rotated mirror image there is always a single axis to flip it about to get it back to the original image – all the shapes only required one flip in the mind's eye. How it determines where this axis is was not explained...
So in conclusion, the world you see is not the actual world – it is blemish corrected, your blind spot (where the optic nerve comes out of the retina) is filled in with guesswork, and is heavily image processed. Our heavy reliance on vision has shaped our thoughts and metaphors in our language.
And yes, that is my eye.