Modern smartphones shape how our memories will look — from simple enhancements to fantastic facial features. The results may please you, but they can change our perception of the surrounding world, claims the BBC.
When you tried to photograph the Moon with your smartphone, you probably felt disappointment unless you had a Samsung Galaxy. These devices boast a "100x space zoom" feature that allows for surprisingly clear shots of the Moon, despite the modest size of the lens. Such a level of detail is not available on Apple's iPhone, but there is one caveat: photos of the Moon taken on Samsung turn out to be fakes.
One Reddit user shared their experience trying to photograph a deliberately blurred and pixelated Moon from a computer screen. The phone easily captured a clear shot showing craters and shadows that were absent from the original image. Samsung calls this a "detail enhancement feature," but in reality, the company trained artificial intelligence to recognize the Moon and add missing details when the camera cannot capture them.
Not every phone model has such an impressive feature by default. However, regardless of what gadget you have, every press of the camera button triggers complex algorithms and AI-based data processing tools that work in the background. These systems can perform trillions of operations before the image is saved in your gallery.
Ultimately, all these technologies aim to create beautiful and, in most cases, accurate images. However, in rare situations, some phones apply AI-based enhancements that can differ significantly from what you see with your own eyes. The next time you take a photo, consider: is your camera capturing reality, or is it trying to create an alternative version of it?
“This is computational photography,” explains Ziv Attar, CEO of Glass Imaging, who was involved in developing the portrait mode for the iPhone. “Your phone does much more than just capture the light hitting the camera sensors. It tries to predict what the image might look like if the camera were more advanced and recreates it for you,” he adds.
“AI-based algorithms are developed to enhance image quality while maintaining authenticity,” comments a Samsung representative. “Users can turn off AI features according to their individual preferences,” he adds.
However, even if you disable the AI editing features that attract media attention, the algorithms still continue to process the photos you take.
What happens when you shoot?
“When you press the ‘Capture’ button, your phone doesn’t take a single shot; it usually takes between four to ten shots in good lighting,” says Attar. These images are then combined to create a higher-quality shot than if only one frame were used. Some of these photos may be duplicates, while others highlight different elements of the image.
These and other basic processes enhance the photograph by eliminating issues that an ordinary person is likely to overlook. For example, noise reduction minimizes random errors that manifest as graininess. Color correction helps bring the image closer to what you see in real life. There is also High Dynamic Range (HDR), which combines several photos taken under different lighting conditions to retain details in both shadows and highlights. Your phone also actively combats blurriness by applying multi-layer processing methods.
For instance, the iPhone uses a feature called Deep Fusion — an AI-based technology trained on millions of images. It not only processes many of the aforementioned methods but can also identify objects in the shots and process them differently, altering individual pixels based on previously studied images. “This is a very high level of segmentation,” says Attar.
As a result, you get clear and vibrant photos in good lighting conditions. However, some critics and attentive photography enthusiasts argue that modern smartphones go too far and create images with strange, plastic textures or flat textures reminiscent of watercolor paintings. Phones work so hard to eliminate issues that they can even create odd distortions similar to AI hallucinations when zooming in on fine details. Some users are so frustrated with the overly edited photos produced by new phones that they revert to older models or use a second phone solely for photography.
“At Apple, we have always aimed to help users capture real moments so they can relive their memories,” says a company representative. “While we see tremendous potential in artificial intelligence, we also highly value the traditions of photography and believe it should be treated with respect. We remain focused on creating devices that produce genuine and authentic images that look great, and we provide users with tools to personalize them as they wish.”
On the other hand, this can also be viewed positively. Previously, most editing had to be done manually if you were skilled and patient enough. Now, “instead of fiddling with all these different settings, we have automation,” says Lev Manovich, a professor of digital culture and media at the City University of New York. “Some features that were previously available only to professionals are now accessible to amateurs.”
Nevertheless, your phone often takes on creative or even artistic decisions regarding the moments you capture. Users may not be aware of this, and in some devices, AI does much more than just adjust settings.
“I believe smartphone manufacturers genuinely want photographs to reflect what people captured. They are not trying to create fakes,” says Rafal Mantiuk, a professor of graphics and displays at the University of Cambridge in the UK. “However, there is a lot of creative freedom in the image processing. Each phone has its unique style, you know? Pixel phones have their style, Apple has its own. It’s like the work of different photographers.”
“This is pure hallucination”
In these debates lies an implicit standard: the notion that a “real” photograph should look as if it were taken in the era of film photography. This comparison is hardly accurate. Every camera has included certain image processing techniques from the very beginning. The term “AI” often evokes negative associations. However, in most cases, algorithms help correct shortcomings associated with the tiny lenses and sensors used in phone cameras.
Nevertheless, some features go beyond simple enhancements.
