If you’ve visited a newsstand in the past decade or so, then you know it’s impossible to escape digital retouching — and we’re not just talking about the covers of fashion magazines. Tired of all of the “impossibly thin, tall, and wrinkle- and blemish-free models” overrunning the media, researchers Hany Farid and Eric Kee of Dartmouth College have created a new algorithm that mimics human perceptions and can detect when a model’s face has been retouched using Photoshop with 80% accuracy. “The ubiquity of these unrealistic and highly idealized images has been linked to eating disorders and body-image dissatisfaction in men, women, and children,” they write in The Proceedings of the National Academy of Sciences. They believe that if viewers know how much an image has been altered, it might help.
Their proposal? A 1-to-5 manipulation “rating system” for photos that will reveal both geometric adjustments (things like slimming the hips or enlarging the eyes) and photometric adjustments (things like removing wrinkles or cellulite) that have been made to retouched photos. “Such a rating may provide incentive for publishers and models to reduce some of the more extreme forms of digital retouching that are common today,” they explain. “It remains to be seen if this rating can mediate the adverse effects of being inundated with unrealistic body images.”
While we can’t imagine that publishers or models exactly going for this, as this kind of software becomes more readily available, will they really have a choice? We’re also curious: Do you agree that photo-labeling can counteract the negative effects of manipulated images? Do you think it will create a greater sense of restraint within the industry? [via NYT]