An Indian-American student at Massachusetts Institute of Technology, US, has devised a formula that tells how the contents of a photograph may predict its popularity online. (Agencies)
Computer student Aditya Khosla and his team scanned through 2.3 million Flicker photos to see which got the most views.
They then looked for correlations between the colour, composition and subject of an image and that picture's likelihood of missing 'likes'.
They found that "brassieres", "revolvers", "miniskirts", "bikinis" and "cups" have a strong positive impact on a photo's popularity.
"Some people have 10 friends, some have a thousand. Despite all these differences, it is interesting to see that the content of the image itself can be used to predict how popular an image is going to be," Khosla was quoted as saying in a leading website report.
While underwear and dinnerware get more 'likes', objects like laptops, golf carts and space heaters had no takers.
Khosla suggested while posting a photo on social media, junk greenish and blue-gray hues as these colours "tend to be less popular".
The researchers' algorithm suggests colours like aqua, bright red, navy and chartreuse.
"Open scenes with little activity tend to be unpopular," he said.
The new findings can help create software that would let users edit their photos to make them more appealing, the report added.
An Indian-American student at Massachusetts Institute of Technology, US, has devised a formula that tells how the contents of a photograph may predict its popularity online.