Having “solved” the wrinkle problem, camera apps also come up with teeth whitening solutions

Along with satellite connectivity, ray-tracing support and faster processing cores, new smartphone chipsets also introduce more advanced filters for processing camera images, which will become part of the standard arsenal offered on many manufacturers’ devices.

Qualcomm and MediaTek have been announcing their new generation chipsets since late last year, but the benefits of the processing power brought to phones like the OnePlus 11, Xiaomi 13 and Vivo X90 are only now being revealed.

Until now, features such as eye shrinking/enlargement and fundamentally altering facial geometry have been “innovations” reserved mostly for consumers in China. In contrast, retouching to the complete removal of expression lines has been more of a failed experiment for Asian manufacturers who have pre-activated such filters on devices shipped to consumers in Western markets.

Somewhat more grounded on the hardware side, the next experiment was a megapixel race, bringing cameras of up to 200MP to new generations of smartphones. But these would not be possible without increasingly sophisticated image processing technologies. For example, artificial intelligence can now divide the image into regions, applying specialised filters in real time to the sky, vegetation, faces and bodies of people in the frame. Instead of the picture being placed in a certain category (e.g. nature) and processed according to a pre-defined profile, the elements in the image are identified and processed separately. This way you can get a pastel sky without the subjects’ faces looking cartoonish, or independently light/darken certain elements in the image.

Read:  Tesla announces its own interpretation of the AirPower charger, which Apple has dropped

But why not apply even more granularity to each person in the frame. For example, compensating for facial features based on skin color, presence or absence of glasses, type of clothing worn, race and gender.

Although, not limited to this, AI processing functions supported on new generations of Qualcomm and MediaTek chipsets can achieve selective and instantaneous teeth whitening, as well as eliminate other imperfections such as reflections appearing in eyeglass lenses. Illustrated in a video distributed on the Snapdragon channel, attributed to the Qualcomm manufacturer, the new automatic filters can make people of colour look much lighter in complexion without ruining the tonality of Asian or European faces captured in the same frame.

Read:  Cryptocurrency "miners" have unlocked NVIDIA LHR video cards

Many smartphone brands use semantic segmentation for single-camera portrait modes, while other brands use it for AI scene recognition (sunsets, landscapes, flowers, food). Taking things even further, some manufacturers like Xiaomi and Google are promoting the ability to fundamentally alter the look of the image, swapping a cloudy sky for a completely clear landscape.

Qualcomm goes one step further, promoting for new generations of chipsets the ability to identify specific elements such as teeth, hair, facial hair, lips, fabric and more. And that could open the door to exciting new possibilities, with the camera app offering instant teeth whitening, facial hair removal, and even graying hair “repair”.

The Best Online Bookmakers October 03 2024

BetMGM Casino

BetMGM Casino

Bonus

$1,000