Distraction-predicting software in action; photos on the right are the ones with distractions [in red] removed.
Fried et al, “Finding Distractors In Images”
In an effort to help people take “better” photos, a team of computer scientists from Princeton University and software company Adobe have created a program to automatically remove distracting elements from your photos.
Their software goes one step further than Photoshop’s magic healing or clone brush: with just a single tap, their software automatically identify and remove the distracting elements, which can include objects near the border of the photo, a face coming in from the side (assumed to be a photobomb), a piece of trash on the lawn, a person peeking from atop a structure, the side of a table peeking out in the corner, a vertical window, bokeh light effects, etc. In other words, elements that may be important to the photographer.
To determine what elements may be distracting, they used volunteers, found through Amazon’s Mechanical Turk system and through the photo editing app Fixel, to sift through thousands of photos to annotate or edit out the distractors. I am surprised not one of them mentioned distracting power lines.
They hope to incorporate the algorithm into a photo editing software. Hopefully, not directly into your camera.
