Princess Eye Area

When I saw Tangled, I noticed how large the princess’ eyes were.

Then Frozen came out, and their eyes were equally as large, it seemed. I figured that the new generation of CG animators favored large eyes as the favored style of beauty. But I also wondered if this could be quantified.

So I set out to objectively quantify it. It being the size of each Disney princess’ eyes relative to her face, as a percentage.

Here is my conclusion:

No, the relative eye sizes are not increasing. Rather, it is the wider head shape that makes it seem that way.

The winner is actually Tiana.

Here are the data to support that:

graph of Disney princess eye area as a percentage of facial area versus time

Princess Year View Eye Pixels Face Pixels Percentage
Snow White 1937 front 61 1554 3.925
side 21 1222 1.718
Cinderella 1950 front 45 1285 3.501
side 17 1584 1.073
Aurora 1959 front 53 1256 4.219
side 23 1174 1.959
Ariel 1989 front 55 968 5.681
side 48 1019 4.710*
Belle 1991 front 99 1525 6.491
side 24 928 2.586
Jasmine 1992 front 110 1581 6.957
side 37 1277 2.897
Tiana 2009 front 144 1878 7.667
side 26 1253 2.075
Rapunzel 2010 front 120 1788 6.711
side 26 1050 2.476
Anna 2013 front 82 1265 6.482
side 16 795 2.0125

Note: while I wanted this to be an objective measurement, hair complicated things. Since the size and shape of hair and hairstyle varied from princess to princess, it needed to be eliminated from the data. So there is still a fair amount of subjective evaluation. I had to crop each image to the face (for front images) or head (for profile images) of each princess. For many, since the hair was significant, I had to guess where the edge of the face or outline of the skull was. My reference would be better termed “perceived facial area” or “perceived head size”.

* I didn’t have a good profile of Ariel’s face, so don’t trust this number.


How did I go about this? Here’s a walkthrough of my method, using Anna as a representative example.

First, I collected front and profile images of each princess.

image of Princess Anna from Disney's Frozen

Next, I normalized them. I chose a size of about 50×50 pixels, mainly because I wanted an easier time counting pixels. I cropped the image close around their faces so as to maximize the facial area in the image.

image of Princess Anna from Disney's Frozen

Then, I started counting the number of pixels in each eye.

After that, I realized how many pixels I would have to count to get an idea of the size of each face, and I decided not to count them manually.

After a quick search on the internet, I decided that the histogram would be the best way to go. So I erased the portion of the image that was not her face.

image of just the face of Princess Anna from Disney's Frozen

The Value option of the histogram tells how many pixels there are with any sort of value. Erased pixels have no value, so this told me how many pixels there were of her face.

image of histogram for the face of Princess Anna from Disney's Frozen

1265

Then, I erased everything but her eyes.

image of just the eyes of Princess Anna from Disney's Frozen

The histogram then told me the area of her eyes, in pixels.

image of histogram for the face of Princess Anna from Disney's Frozen

82

Then I noted the pixels counts and divided the eyes by the face to get the percentage.

82/1265 = 6.482%

Repeat for profile, then keep repeating for all princesses.

And Leah’s eyes were weak, but Rachel was beautiful of form and face.

Genesis 29:17

Digg Del.icio.us Reddit Stumble Upon

This little article thingy was written by Some Guy sometime around 6:21 am and has been carefully placed in the Ponder category.

Leave a Reply

Comment moderation: please do not submit your comment multiple times, as comments are not posted until I approve them. If your comment never appears, that probably means that I didn't like your comment (maybe off topic, maybe spam, maybe not family-friendly, etc.).

Check this box to be notified of additional comments to this post. Notification will be sent to the email address you provided above. You can also subscribe without commenting.