Do we see objects as they really are or does our brain manipulates itself? The apparent size of an object decreases with increase in distance that's why we see rails get closer as they  get farther. We don't see the rails parallel throughout their length but converging. But there is a contradiction:
Let,
the real size of the object be s, 
the distance between the eye and the object be d, 
the visual angle be Ɵ, 
let the retinal image size be I
the radius of eye ball is r = 8.5 mm

From the above figure, the visual angle is

and the arc length I = 2Ɵ*r;

Here r and s are constants. Thus we get a relation between retinal image size and distance (between the eye and the object).
A graph for I vs d (for s=1) is shown below,

The retinal image size doesn’t varies directly with distance. The values of I for certain d’s are given below,
d=0.5 I= 0.05340708
d=1, I= 0.03152804
d=2, I= 0.03152804
d=3 , I= 0.06305608
d=4 , I= 0.00845614
d=5 , I= 0.006777468
d=10, I= 0.00339717
and here is an illustration of how things might work.

Is the above explanation correct?
 A: The interplay between the retina and the brain is incredibly complex.
In your example, you can't be focused both near and far at the same time, so the brain "fills in" the information as it changes the focus. And where we expect straight lines, we will see straight lines.
A lot of the processing in the visual system is differential: we compare with other things nearby. This is why a series of steps in grey scale look like the individual boxes have a gradient in them: at one end we see "darker than the next box", at the other end it's "lighter than the next box", so we decide there must be a gradient that isn't there.
From http://www.spearsandmunsil.com/cms/wp-content/uploads/2013/02/1920x1080p24_11-Step_Crossed_Gray_Steps_Main.png:

Similarly, we have all kinds of blood vessels in front of the retina (really - what terrible design decision that was. Only the octopus got it right, IIRC.) - and even a whopping big hole (the blind spot) where the optic nerve goes through the retina. But we don't "notice".
So yeah - whatever is projected on the retina is only loosely related to what we "see". There are countless optical illusions designed to exploit the difference.
