Shopping? Check out our latest product comparisons

Proposed rating system would indicate how much photos have been retouched

By

November 29, 2011

On a scale of 1 to 5, how much retouching would you say was applied to the left-hand image...

On a scale of 1 to 5, how much retouching would you say was applied to the left-hand image of Kim Cattrall, to arrive at the vision of loveliness on the right?

Even though we know that the photos we see of models and celebrities are retouched, many of us nonetheless can't help but think "Yeah, but even without that little bit of airbrushing, that person still looks way better than me." For most people, such thoughts are merely a little bit humbling. For others, however, they can lead to Body Dysmorphic Disorder, eating disorders, or severely-low self-esteem - all of which can in turn have very serious consequences, including death. Perhaps if those people knew just how retouched that one photo of Mila Kunis or Ryan Gosling was, however, they might realize how much of a lie it really represented. That's why researchers at New Hampshire's Dartmouth College are proposing a system that objectively assesses the extent of changes made to a photograph, then displays that amount as a number rating on the published photo.

Dartmouth Computer Science Professor Hany Farid and PhD student Eric Kee have created a basic version of the system, which applies a mathematical model to differences between digital "before" and "after" photographs of people. These differences could include changes made via cropping, color adjustment, the slimming or shaping of body parts, the adjustment of facial symmetry, and the removal of features such as wrinkles, blemishes or bags under the eyes.

A number ranging from 1 to 5 is then assigned to each photo, indicating the degree to which it has been altered. In order to check the validity of their system, Farid and Kee got a group of volunteers to compare and rank the differences in hundreds of pairs of before-and-after retouched photographs. The average numbers arrived at by those human test subjects correlated strongly with those generated by the model, for the same images.

The idea is that ultimately, advertisers and other media outlets would be required to display these numbers on their retouched images, so that everyone would know just how unattainable of an ideal was being depicted. According to a report in New Scientist, the Dartmouth system could also be installed as a Photoshop plug-in, which would warn retouchers when they were deviating too far from the original image.

About the Author
Ben Coxworth An experienced freelance writer, videographer and television producer, Ben's interest in all forms of innovation is particularly fanatical when it comes to human-powered transportation, film-making gear, environmentally-friendly technologies and anything that's designed to go underwater. He lives in Edmonton, Alberta, where he spends a lot of time going over the handlebars of his mountain bike, hanging out in off-leash parks, and wishing the Pacific Ocean wasn't so far away.   All articles by Ben Coxworth
4 Comments

This is an absolutely great idea. And the PS plugin as well. Great stuff!

Renārs Grebežs
29th November, 2011 @ 11:32 pm PST

Major point of irritation - trying to make 50+ year old women look 14 to 16 years old - with great tits and wide hips - is one more reason to dump Hollywierd and it's collective fakeness anyway.

Mr Stiffy
30th November, 2011 @ 01:20 am PST

ok but how often do you have the original to score it against?

never

wle

wle
30th November, 2011 @ 08:16 am PST

One of the problems with this story is the attitude expressed in the caption of the example photos of Kim Cattrall. The assumption that the image on the right is a "vision of loveliness" leaves us to understand that the image on the left of a 50+ year old woman is the opposite. A rating system for photo retouching is likely well overdue. What is truly needed to transform our perceptions of "beauty" though may be a much bigger fish to fry.

Joseph DeLappe
30th November, 2011 @ 08:35 am PST
Post a Comment

Login with your gizmag account:

Or Login with Facebook:


Related Articles
Looking for something? Search our 27,809 articles