Make your smartphone selfies even smarter with this algorithm

Even though anyone can run any photo through a filter, MIT researchers wanted a more interesting way to add fabulous features to photos. Developers at MIT teamed up with Adobe Systems and the University of Virginia to establish an algorithm to add flattering styles of different professional photographers to enhance your smartphone-snapped-photos.

This billion-dollar “style transfer” industry produced from image altering apps like Instagram is not necessarily well-matched with close-up facial features when image filters can sometimes distort the selfie. For example, if a person takes a selfie with glasses on in harsh lighting, the camera will reflect light off of the glasses. With this software, the glare would be eliminated and replaced with a modified image with studio lighting. The software developed by the MIT-lead research team replicates the portrait styles of existing photographers like Diane Arbus, Richard Avedon, and Martin Schoeller and their signature vintage styles. For example, Richard Avedon's photography uses unconventional forms of studio lighting, such as strobe lighting. A person's image through the “filter” of Richard Avedon's signature style would then appear with strobe lighting with MIT's algorithm tech. All of the image subject’s facial features are taken into consideration by the algorithm before it is applied to the photo. Robert Bailey from Adobe stated, “You can't get stylizations that are this strong with those kinds of filters. You can take a photo that has relatively flat lighting and bring out portrait-style pro lighting on it and remap the highlights as well.”

This endeavor will be presented at the graphics conference Siggraph this August. Do you think this program will be readily accessible as an app and gain popularity like Instagram in the future?

Story via MIT