In the era of Instagram, we're constantly asking ourselves, "How do I get more likes?"
And, of course, MIT has the answer. (Who else?)
A team at the school's Computer Science and Artificial Intelligence Laboratory have developed an algorithm that can actually predict how memorable or forgettable an image will be. According to the lab, the software can go head-to-head with human judgment — meaning it can tell how much of an impact your photo will have (almost!) as well as you can.
You can actually try out the algorithm for yourself by uploading your own photos here. It'll create a heat map that identifies which parts of the picture are most memorable, like these:
Understanding memorability is important, says CSAIL graduate student Aditya Khosla, who was lead author on a related paper, because it "can help us make systems to capture the most important information, or, conversely, to store information that humans will most likely forget. It’s like having an instant focus group that tells you how likely it is that someone will remember a visual message.”
This technology, she believes, could improve the content of ads and social media posts and help to develop more effective teaching resources.
But how does it work? We'll let the lab explain that:
Neural networks work to correlate data without any human guidance on what the underlying causes or correlations might be. They are organized in layers of processing units that each perform random computations on the data in succession. As the network receives more data, it readjusts to produce more accurate predictions.
The team fed its algorithm tens of thousands of images from several different datasets, including LaMem and the scene-oriented SUN and Places (all of which were developed at CSAIL). The images had each received a "memorability score" based on the ability of human subjects to remember them in online experiments.
The team then pitted its algorithm against human subjects by having the model predicting how memorable a group of people would find a new never-before-seen image. It performed 30 percent better than existing algorithms and was within a few percentage points of the average human performance.
A key takeaway for your next Instagram shot: The system tends to rate wide shots and landscapes as less memorable than images with prominent subjects.
So, yes, that goofy selfie will probably do better than a sunset. And maybe blur out anything distracting in the background because, duh, it's all about you.
Photo Credit: Getty Images