iOS 16 makes it easy to meme your friends, or create YouTube thumbnails
When I tested the developer beta version of iPadOS 16 last week, I gave the bulk of the attention to the new “Stage Manager” feature, which when paired with (also new) external display support, fundamentally changes what an iPad can become. But there was another feature I briefly touched on that has plenty of potential: the ability to quickly isolate and “cut out” subjects/objects from the background of a still photo or a video. Apple did not give this feature a new name, but rather grouped it as part of the “Visual Look Up” feature that’s been available since last year. And yes, this feature is also available on iOS 16.
Cutting out a person or an object from an image as a transparent PNG file usually involves professional software like Adobe Photoshop along with a Wacom stylus. In recent years, smartphone apps like Cut Paste Photos or web-based design services like Canva have offered a quicker way to do this, but the results aren’t as clean and it still takes between a minute or two of fiddling to get right.
What makes Apple’s new background-removing feature so uncanny is the fact that it also works for videos, and also that the cutting process takes virtually no time at all. You just have to view the photo or video in Apple’s Photos app, long press on the subject/object you want to isolate for half a second, and Apple’s machine learning-based object recognition algorithm will isolate the subject/object almost instantly. After this, you can share the isolated PNG image in three ways: you can “copy” it (as if you’re copying text) and paste it in another window; or tap on the “share” button and select another app to bring the image there; or you can long-press the image again after it’s been cut and hold on to it while you move to another app and drag it there. This last drag-and-drop method was probably designed for iPadOS 16, but it works on iOS 16 too.
If you need a more step-by-step guide on this, check out our guide on how to remove the background from images on iOS 16.
When Apple demoed this photo cutout feature during its WWDC keynote, it showed the image being shared to Apple’s native Message app, which was to be expected.
But when I got to test the feature for myself, I was pleasantly surprised (and very excited) to see that this feature also works for third-party apps such as LumaFusion and Pixelmator (in my opinion, iOS’ best video and photo editing app, and better than any counterpart available on the Google Play Store).
Plenty of creative professionals, including myself, need to cut subjects from the background as part of their regular workflow. I make YouTube videos in my free time and my video thumbnails often feature a cutout of a phone (or me holding the phone) pasted over another background or text. I rely on Canva to do this, and it’s a process that takes about three to five minutes. Now, with iOS 16 or iPadOS 16, I can pull off similar tasks in a few seconds. See the example below, when I quickly put together a mock thumbnail by grabbing two cutouts of myself (including one from a video) and pasted onto another image in Pixelmator. The whole process took under 10 seconds, and that’s just beautiful.
Okay, not everyone needs to make a YouTube thumbnail. But almost everyone enjoys memes, right? I see this new cutout feature can significantly level up our collective meme game. How could it not? Anyone with a recent iPhone or iPad will soon be able to quickly grab a cutout of their friends or family members, and just use it to create hyper-contextual memes in the moment, without really needing specialized tools or any technical skill.
For example, my girlfriend dabbles in Thai boxing, and the image of her soft-spoken, gentle self trying to hit things is amusing to me, so I quickly made the below image of her as a joke. Again, the process of cutting it out, pasting the cutout to another image, then sending it via iMessage took under 10 seconds.
And this cutout feature works for virtually any photo and video, as long as it’s saved to your iPhone/iPad’s Photo app. Even a low-res scan from a 1996 issue of a basketball magazine works. Below, I cut out the NBA players from the cover, and a photo of my girlfriend and me with a dog, and quickly slapped them together in iPhone’s Notes app.
Or, in the sample below, I pasted a cutout of my friend (having fun at a wedding reception) from a video onto a photo of male exotic dancers and shared it via Telegram with another friend. (Note: I have my friend’s permission to use his photo as meme tests)
The meme possibilities are endless, as long as you have a subject and an iPhone. Sure, we could do all this before as well, but it required a third-party app (and you usually have to pay for one that works well) and five to ten minutes of fiddling. Now you can do all this in under 10 seconds — it’s the spontaneity of it all that makes it so fun to play with. I can snap a photo of a friend at a bar, notice his facial expression is funny, and immediately make a meme right there on the spot.
And considering that this is a first developer beta (not meant for public use), the feature is already working extremely well. The only thing I notice that Apple needs to address is the long-press action to begin cutout sometimes will trigger Live Photos instead (Live Photos has historically been triggered on iOS via a long-press too). Right now, the solution seems to be a short long press for the cutout, then a longer long press for Live Photos to begin playing. You can see how this process can be complicated for some. Perhaps Apple will have to assign Live Photos playback to another action down the line. Wouldn’t it be great if iPhones still supported 3D Touch?
But on a serious note, this new cutout feature in Visual Look Up is an impressive flex of machine learning prowess by Apple’s silicon. The fact that it can identify a person or a thing from a still photo or video almost instantaneously is wild, and I just can’t picture any other mobile device save for a Google Pixel which could feasibly do this within the next year or two.