Due to various issues my eyes don't work together (I look out of one of the other depending on a mix of focal length and just what my brain randomly choses), but we get 3D information from more than just using two eyes in unison. The parallax information that two eyes allows us to infer can also be derived by movement, either of objects relative to each other or by our head movements. Ever watch a cat or other animal lift its head up a bit then resetting prior to a jump/pounce/other, even if the view is clear so they don't need to look over something? That is partly what they are doing, gaining extra accuracy by being able to do parallax computation based on information from four points (both eyes at different heights) instead of two.