Google‘s new Google Pixel 9 phones are full of useful Gemini AI-supported skills, including the startlingly capable Add Me group photograph Camera App utility that lets you add yourself to a group photo, even if you’re the one taking the picture.
I think that’s grand, but I couldn’t help wondering if this powerful Gemini AI feature could be employed for some less utilitarian activities, like copying myself.
Prior to getting my hands on a Google Pixel 9 review unit, I tried Add Me with the assistance of some Google representatives, The operation appeared simple. There’s now an “Add me” option in the camera app. Once selected, on-screen guidance tells you to “Move the camera to scan area.”
It’s a quick process; you know you’re done when the shutter button turns entirely white. After that, you can take a photo of your group, but be sure to leave enough room for future you to be added. To add yourself, you tell everyone in the original shot to step out and hand the camera to someone else. You do have to make sure they don’t move the camera very far.
The viewfinder shows an augmented-reality version of the scene with a semitranslucent version of the original group still there. Whoever is holding the Pixel 9 uses that to guide where you should stand in the final photo without overlapping the original group.
Once you’ve taken that second photo, the system takes five or six seconds to combine the two images into a rather perfectly stitched whole.
I loved this feature from the minute I saw it and wondered about its flexibility. First of all, I was at home and didn’t have access to a group of people for an Add Me tryout. What if I made a group of myself?
Dupe yourself
Initially, I thought all I would need to do is set the camera timer between shots and, Bob’s your uncle, there are multiple Lances. For all its power, though, Add me is a somewhat limited photographic feature. There’s no ability to set a timer at any point. I needed to figure out how to take two photos of me without holding the camera.
I put the Google Pixel 9 on a tripod and positioned it in front of my couch, placing it near enough that I could reach around the phone and find the shutter button (no mean feat since this is a screen and not a real button). Luckily, my arms are long enough that I could sit on the couch, fit myself into the frame, and reach the phone’s screen that was facing away from me.
Image 1 of 3
I started each shot by moving the phone around to scan the area, then positioned myself on the far right of the frame, reached, and took the first shot. If this were a normal Add Me shot, I’d use the Pixel 9 viewfinder screen to position the second me in the frame. Instead, I just positioned myself to the far left of the frame (knowing I couldn’t move too far since this was a pretty tight shot), reached for the Pixel 9 screen, and tapped until I located the shutter button.
When I circled back around to view the final shot, Pixel 9’s Add Me had stitched together a “twins” photo of me. While it initially took some trial and error (I took more than a few shots of just one me and a couple where no one was in the shot), I got pretty good at it and ended up doing a few Lance clone photos.
More friends
Dual selfies are not the only thing possible with Add Me. You can make dual portraits of friends and colleagues. To grab this shot of TechRadar Managing Editor Jake Krol, I had him sit in front of me while I held the camera still. For the second shot, I asked him to slide his chair to the left while I lined up the second him with the original AR version of Jake. Once again, the results were essentially perfect. I think they’re more entertaining when you have a different expression or action in the individual portraits, especially if the end result makes them look like they’re aware of each other.
Add Me has a couple of notable limitations. You can’t take more than two photos for each Add Me shot, so adding more than one me was out of the question. Add Me only works with people. No inanimate objects or, it seems, pets. When I tried it with stuffed animals, cups, and other objects, Add Me never created the AR version of the original shot. It’s not a big deal and I do wonder if Google might consider extending Add Me to accommodate these kinds of photo subjects.
Even so, it’s gratifying to discover a use for one Gemini AI feature that is not all about utility or simplicity and is, instead, about a little bit of Multiplicity.