Google’s Pixel phones have always had stand-out cameras, thanks in no small part to the company’s use of AI. The latest application? A built-in kiss cam for the camera on the Pixel 3 that automatically detects when subjects are puckering up and snaps a quick photo.
The feature is an update the Pixel Camera app’s Photobooth mode. This is a shutter-free mode which automatically takes photos with the Pixel 3’s wide-angle selfie cam. In addition to spotting kisses, Google says the software recognizes five key facial expressions that “should” trigger capture: “smiles, tongue-out, kissy/duck face, puffy-cheeks, and surprise.”
That’s the theory, anyway. Our tests with the app were inconsistent. “Its ability to detect duck-face is questionable,” was the assessment of The Verge’s Jon Porter. Though he added it did successfully spot him kissing his reflection in a mirror. “It grabbed the image the moment my lips made contact!”
The tech for this comes in part from Google Clips, the company’s 2017 experiment in using AI to make photography easier. Clips was supposed to be a tool for families to capture important moments. It was small, lightweight, and minimalist, and used built-in algorithms to decide when to take a photo. But while a neat concept, it was redundant for most users.
While Clips has been clipped out of Google’s history (we couldn’t find it for sale on the Google store), the tech it helped incubate lives on. With neural nets scanning your facial expressions and making sure your eyes aren’t closed, Google says the Pixel 3 makes it easier than ever to take perfect selfies and group photos.
do you trust artificial intelligence to take a good photo?
As part of the Pixel Camera’s update, the app also helps users know when they’re looking their best for a photo. A white bar on the side of the display (on the left in the GIF above) responds to users’ actions. When everyone’s looking at the camera and making a nice face it expands to the full width of the display and the phone takes a picture.
“We’re excited by the possibilities of automatic photography on camera phones,” write Google’s engineers in a blog post. “As computer vision continues to improve, in the future we may generally trust smart cameras to select a great moment to capture.”