Best ever optical illusion

Yeeek said:
Okay i must have weird eyes because i don't see them with the same color.. :(
Your eyes are perfectly normal. The shade of the cylinder makes the patern darker... and your brain interpret that it's a shade and tend to compensate so that you recognize the pattern as one and the same.

However, if you focus on the grey of A square, you'll see all the areas in the same grey better, including the B square. Actually, there are only 6 squares which are in that specific grey :
- A
- the one at the left of A
- B
- the one at the left of B
- the two darkest squares at the bottom of the picture.

EDIT : I had to check on Paintshop to be sure for the last two bottom ones.
 
My computer must have a really crap paint program
 
Eran of Arcadia said:
The two faces are clearly the same face (of a guy) flipped around. Still, I found myself thinking that the right one must be more feminine. Because it had a different eyebrow raised, maybe?

close, but no cigar.
 
kingjoshi said:
close, but no cigar.
Well they are mirror images correct?
I can't see much difference after staring at them for a while. Is one fractionaly lighter in shade than the other?
 
I am quite fond of this subject.Especially St.Augustine argument on the boar being bent in the water.

Here is some of my fav:


The famous bunny or duck illustration:


These last three are somewhat downright sheer manipulation.



 
They're not mirror images. And almost everyone finds the right image more feminine. I wonder if anyone can figure out why.
 
I've always liked the simple 'ol blind spot related illusions

blindspot1bw.gif
 
Wow, that's crazy... (the first one, of course)

As for the faces, the right one does look more feminine to me, too. I examined them closely, and I think I feel this way because the face on the left seems to have slightly bigger lips and a slightly more flattened nose slightly closer together. But it's hard to tell for sure.
 
toh6wy said:
Wow, that's crazy... (the first one, of course)

As for the faces, the right one does look more feminine to me, too. I examined them closely, and I think I feel this way because the face on the left seems to have slightly bigger lips and a slightly more flattened nose slightly closer together. But it's hard to tell for sure.

Actually, they are both blended images. The first picture's left part is a man and right part a woman. And in the second image, it's the opposite section of the SAME man and woman.

But for whatever reason, humans use the left part of a face to determine gender.

Here is another picture. The bottom left is the man and the bottom right is the woman. The top two are the same composites as before.


EDIT: though a lot of people say the guy looks feminine to begin with :p
 
kingjoshi said:
EDIT: though a lot of people say the guy looks feminine to begin with :p
He doesn't seem feminine to me.I think this illustration photo is just something to demostrate the flaw of this so-called "Human and Machine Vision course ."

What was the intent of your professor making this somewhat comparison?I don't see optical illusion anywhere in this.
 
It's not an optical illusion, but I think it's in line with the thread about our eyes. Such a discrepency between "seeing" and "perceiving" and the latter is all the processes our brain does to make sense of what we "see". Part of that is the distribution of processing periphery data versus "focusing" on certain points in an image. Part of that is willingness of the brain to make mistakes to get information after processing data to the other parts of the brain to allow making decisions and so forth.

When looking at an image, how briefly can one display the image and the viewer still guess the gender of the person? Race? Tests have shown that when doing such experiments (white person holding gun, black person holding gun, innocent white bystander, innocent black bystander), that people are 3 times more likely to assume the black bystander is a criminal.

When looking at this image, without even knowing why, our brain can recognize certain differences and uses that to determine gender. And it's beyond any consciousness. She was messing around and came up with this. I don't know her full study, she wasn't complete and I never got the paper.

It's was a grad course and a blast. I think the professor is good enough because after one year at Michigan State, she went on to teach at MIT.
 
kingjoshi said:
It's not an optical illusion, but I think it's in line with the thread about our eyes. Such a discrepency between "seeing" and "perceiving" and the latter is all the processes our brain does to make sense of what we "see". Part of that is the distribution of processing periphery data versus "focusing" on certain points in an image. Part of that is willingness of the brain to make mistakes to get information after processing data to the other parts of the brain to allow making decisions and so forth.

When looking at an image, how briefly can one display the image and the viewer still guess the gender of the person? Race? Tests have shown that when doing such experiments (white person holding gun, black person holding gun, innocent white bystander, innocent black bystander), that people are 3 times more likely to assume the black bystander is a criminal.

When looking at this image, without even knowing why, our brain can recognize certain differences and uses that to determine gender. And it's beyond any consciousness. She was messing around and came up with this. I don't know her full study, she wasn't complete and I never got the paper.

It's was a grad course and a blast. I think the professor is good enough because after one year at Michigan State, she went on to teach at MIT.
Then Michigan State,along side other midwestern universities (which is common in my area from what i gathered by university students here)have a tendency to lose some interesting people to MIT.Hmm...I find it fascinating,too bad that you are not still in her class which does not favor me,cause i was looking forward of a free lecture from via you from the teacher:mischief: .She must made a good impression to you.
 
kingjoshi said:
But for whatever reason, humans use the left part of a face to determine gender.

Hmm. In that case, I'd guess that whatever part of the brain is used to recognize faces is in the right hemisphere.
 
toh6wy said:
Hmm. In that case, I'd guess that whatever part of the brain is used to recognize faces is in the right hemisphere.

This assumes we need to disambiguate information. If there is overwhelming data and you cover you right eye, I'm sure your left hemisphere of the brain can identify the gender :) Humans have quite a lot of redundancy.

CartesianFart said:
Then Michigan State,along side other midwestern universities (which is common in my area from what i gathered by university students here)have a tendency to lose some interesting people to MIT.Hmm...I find it fascinating,too bad that you are not still in her class which does not favor me,cause i was looking forward of a free lecture from via you from the teacher:mischief: .She must made a good impression to you.

Yes she did. She was French and had a great accent! :love:

But anyway... A lot of lectures from MIT are online. They post course material and even videos. It's part of their "OpenCourseWare" program, similar to "Open Source". They believe the distinguishing part of MIT is being in the environment with other intelligent people and professors (as well as massive funding :p).

She's teaches unter the Brain and Cognitives Sciences program. And they also have classes in OpenCourseWare.
 
Yeeek said:
Okay i must have weird eyes because i don't see them with the same color.. :(
I'm with Yeeek, nothing I do makes them the same. I assume we are correct and all the rest of you are delusional. I'll ask my wife and see what she says. She knows the truth when she sees it. ;)
 
Birdjaguar said:
I'm with Yeeek, nothing I do makes them the same. I assume we are correct and all the rest of you are delusional. I'll ask my wife and see what she says. She knows the truth when she sees it. ;)

Two things you can do.

1) open up the image in some program that allows you to select a color based upon a pixel. Go to a pixel in square A and square B. They're the same color.

2) open up the image in a program like MS Paint. Just erase all the other stuff until only two those squares are left. then you'll see they're the same color.
 
Back
Top Bottom