Computer scientists at the Massachusetts Institute of Technology have developed a new technique to help selfie takers and online daters post pictures on social media that are more memorable to their audience.

According to The Christian Science Monitor, scientists at the institute's Computer Science and Artificial Intelligence Laboratory (CSAIL) have programmed computers in a way that will guide users to post the most memorable pictures.

Aditya Khosla, a PhD student in computer science at MIT, told The Christian Science Monitor that though algorithms already existed to enable computers to predict information --- like the auto-complete function in Google search, scientists till now had not been able to use these tools to teach computers to predict what people will remember.

CSAIL principal research scientist Aude Olivia said in a MIT press release that predicting human memory had largely been viewed as a higher-level cognitive process that computer scientists will never be able to tackle. "Well, we can, and we did!" says Olivia.

The research involved testing the algorithm on 5,000 participants worldwide, using a crowdsourcing site called Mechanical Turk, which is an Amazon company that connects developers with people who have signed up to do tasks for small payments.

Participants were shown hundreds of images, each of them displayed for just 600 milliseconds. The researchers made a record of photos that people were able to identify after seeing them only once. These images were then fed into the algorithm to test if it was possible to identify the “memorable” ones.

It was found that the algorithm was good at identifying patterns among different images featuring different scenes including human faces and pictures of nature.

For each photo, the “MemNet” algorithm — which one can try out online by uploading their own photos — also creates a heat map that identifies exactly which parts of the image are most memorable.

This is a significant development which could, one day, help tell people which of their selfies or portraits is likely to get the most likes on Instagram, or to attract more suitors on a dating site.

The science behind the memorability of images could also be applied beyond social media popularity, such as to improve advertising images and those used for educational purposes, according to the Design & Trend.

Contact the writer at feedback@ibtimes.com.au or tell us what you think below.

RELATED STORIES