Google has integrated a new artificial intelligence functionality into its Gemini app, designed to transform ordinary selfies into professional-looking portraits. The tool, known as “photo studio”, uses advanced algorithms to generate high-quality images in different styles, without the need for photographic equipment.
The new feature is gradually being made available to app users and seeks to meet the growing demand for improved profile images for social networks and corporate platforms. The technology is accessible directly from mobile devices that have the Gemini application installed.
The process simplifies the creation of personalized images, allowing anyone with a smartphone to access the technology. The tool automates lighting, framing and retouching tasks that would normally require complex editing software or hiring a photographer.
Como the new tool works
The feature operates by sending a single photo of the user, which serves as the basis for artificial intelligence. Após uploading a clear, well-lit selfie, the user can select from different predefined styles or describe the type of image they want. The AI then processes the original image, keeping the person’s facial features, but placing them in new scenarios, with different clothes and studio lighting. The system generates a set of options for the user to choose the one that best meets their needs, offering variations in poses and facial expressions.
Estilos available for creation
The platform offers a diverse range of styles for the generated portraits. Users can opt for more formal looks, ideal for professional profiles.
Outras options include more creative and artistic styles, which alter backgrounds, clothing, and the overall aesthetic of the image.
The tool also allows the combination of elements to generate unique results, adapting to different purposes, from photos for documents to avatars for social networks.
Tecnologia imaging
The basis of “photo studio” functionality is diffusion models, an advanced category of generative artificial intelligence. Esses models are trained with a vast database of images to learn how to create new images from text descriptions or, in this case, a reference image.
Eles work by “deconstructing” and “reconstructing” pixels to generate a result that matches the user’s facial features with the requested style. The technology is capable of recreating skin, hair and clothing textures, as well as simulating complex lighting conditions, resulting in portraits with a high degree of realism.
Principais resource applications
The main application of the feature is creating profile photos for platforms like LinkedIn.
Outro u