MIT, KHOSLA ET AL.Standing out in a crowd—and particularly, the crowded world of social media—may become a little easier as MIT researchers develop a clever way to tweak profile photos to play up a person’s most memorable facial features. Aditya Khosla and others at MIT have found a way to make photographs of faces more memorable or more forgettable, opening new doors in the understanding of memory.
The process was detailed in work presented at the International Conference on Computer Vision in Sydney in December 2013 and earned Khosla a Facebook Graduate Fellowship for further research. According to the researchers, such “feature tuning”—which could be extended from memorability to other qualities, like confidence or trustworthiness—has the potential to change everything from one’s profile photo on Facebook to online dating to political campaigns and advertising strategies.
In Khosla’s first few weeks as a graduate student at MIT, the talk on campus was about bringing psychology together with machine learning—training a computer to identify patterns in data. Aude Oliva, a researcher with MIT’s computer vision and graphics group, explored what makes images memorable and came to an unlikely conclusion.
“They found that all images ...
