Representation
AI is all rage these days. It has become an integral part of our smartphone and we use it in many tasks – typing, talk, search – but it really started with photography.
Big keywords were computational photography and image algorithms, and at some point, each manufacturer began boasting about its new AI eraser that could remove distractions from photos of your vacation. It was mostly the first with the Google Magic Eraser, but now the industry is caught, and we have dared to say that it has surpassed the pixel to remove the AI OBJECT BUJECT.
Being naturally strange, especially when it comes to technology, we decided to set up various AI erasers against each other, but to see if they are worth using.
After a lot of testing around the Office Fiss, we selected four competitors as our test topics – the iPhone 16E with Samsung Galaxy S25 Ultra, Apple Pale Intelligence using Galaxy AI, Google Pixel 9 Pro XL with Gemini AI, and Ziomi 15.
These devices represent the best efforts of their respective companies, and you can expect similar results from their other high-final phones. Generally, midranges do not even do devices – maybe because of their more limited hardware.
Samsung Galaxy S 25 Ultra • Apple Pal iphone 16E • Google Pixel 9 Pro XL • Ziomi 15
We spent a lot of time to choose the right photos for the job. We ended with a selection of 20 images captured by smartphones. The comparison is appropriate this way – each phone will use its AI on other phones captured images.
Samples come in two groups – some disruptions or common shots with others that spoil the photo – a typical example you will use the AI -saved Object Burject Eraser. Vista wasted street sign, construction crane between buildings, people block a famous landmark, etc.
Then we have a classic photobber shot shot where someone has crashed on your vacation peak and needs to be removed.
There are hard things to remove in other groups of photos – one hand in front of the face, sunglasses that need to disappear, a bottle of glass cleaner between bottles of souls on the bar – we created these shots to give AI a challenge.
A word on the limits of AI
Before we see the results, a word on each phone’s AI – what it needs to work and how it works.
Samsung
Samsung’s Generative Edit requires both an active Internet connection and a registered Samsung account. It also replaces the manipulated image to 12MP (4,000 x 3,000 px), and adds a watermark to tell you that you were tampered with by AI.
One of the strongest suits of the Galaxy AI is that it does not have restrictions on human beings – that is, it can edit parts of people in the photo – some other competitors cannot yet do it.
Samsung’s solution requires at least tap – open the gallery app, go to the photo, tap once, and you can draw around the barrier you want to remove.
Despite its limits, erasing Samsung’s Galaxy AI OBJECT BUGGUCT is better than the rest. It’s faster – to measure, especially compared to Pixel and Ziomi. Samsung’s solution is also incredible to recognize what you want to remove – draw a loose circle around anything in the photo, and 9 out of 10 times AI will recognize what you want to erase. And mask it perfectly.
Apple
Apple Pal’s clean -up is part of Apple Pal Intelligence, which was recently limited to North America. It has recently spread to more countries and regions, but is still not available in China.
Clean up works on iPhone 16E, iPhone 16, iPhone 16 Plus, iPhone 16 Pro, iPhone 16 Pro Max, iPhone 15 Pro, or iPhone 15 Pro Max; A17 Pro or M1 and any later iPad; Mac, and iOS 18.1, iPados 18.1, or Macos Sequoia 15.1 with M1 or later.
For the first time you set up Apple Pal Intelligence you will only need an active connection to download machine learning, and the offline falline will work later.
You can find the photos app clean, and you will need two taps before you can start drawing circles around the interruptions. The iPhone is not so good to identify what you want to delete – a simple person’s silhouette may require five or six finger drawings for the phone to mask the whole item.
Google’s magic eraser was the first of its kind, but it seems that Google has not improved further since its announcement. It is strange to consider how much airtime the facility is available in Pixel Commercial.
The image editor of Google Photos remains in the Magic Eraser Tools section. That means if you want to remove an OBJECT BUGHCT it is three taps away. Google photos will sometimes find Objects Bjects to automatically remove the photos at ATime Shay and you will ask you to use a magic eraser as a suggestion in the first screen of the image editor, but it just happened shortly after our test.
Magic Eraser does not require an active internet connection. Once you open the photo and choose a magic eraser, he will scan the photo and he thinks he or she is interrupted automatically – this is sometimes in the hands – one of those photos got a crowd of people and we deleted all of them without the need to do anything more.
Google offers a separate pixel studio, which, among other things, can erase the Objects Bjects in the photo. We tested it and saw similar results to the magic eraser tied in the Google photo.
If you need to choose something manually to remove, draw around it and try and choose a magical eraser. It is only average in this regard, requiring multiple inputs to properly mask the human silhouette.
Ziomi
The Ziomino solution lives in the owned gallery and is called AI Eraser, AI Eras Pro or AI Eras 2.0. Before you can finally start drawing around the subjects or objects or objects bugs – it is the most difficult to get a full four tap. It also requires an active internet connection, and it mostly depends on the servers in China because it is sometimes very slow, and even refused to work for a while.
Ziomi’s OBJECT BUJECT BUJECT BUGGER IS NOT GOOD. Drawing around a person got poor results – AI chose only a small part of the full shape, and we needed to draw extra circles to cover a simple human shape – sometimes up to 10.
Choosing multiple OBJECTS bugs is a boring task, and the app will only clean all the masks at a time, and we need to start from scratch.
