Google Bard to adopt Imagen as its image generation tool
Google is expanding its generative AI offerings with an upcoming image generator for Google Bard, as revealed in an unreleased changelog dated January 18.
Following the footsteps of tools like the Search Generative Experience (SGE) and Slides, users will be able to create images by entering a few words that describes the image. This new capability is expected to be powered by Google’s Imagen family of models. Imagen is known for their ability to produce high-quality, photorealistic images.
In December, Google introduced Imagen 2, boasting significantly improved image quality. The technology aims at enterprise use cases, such as logo generation or creating images with text overlays.
Google Bard’s image generator is expected to follow safety features similar to those in SGE. The latter, introduced in October, allows users to “draw” or “sketch” prompts directly into the Search field, generating four images at a time.
Just like SGE, Google Bard will also block results with notable people and explicit content
SGE ensures safety by blocking the generation of harmful, misleading, or explicit content. Bard will also restrict prompts featuring notable people or photorealistic faces, reports 9to5Google. Additionally, each generated image will get a metadata labeling and invisible watermarking using Google’s SynthID.
Google Slides and Meet, the latter for custom backgrounds, already incorporate user-facing image generation through Duet AI. The forthcoming addition of an image generator to Bard extends the accessibility of AI-generated content across various Google services.
By adding further descriptions to image captions in Imagen 2’s training dataset, Google is enhancing image-caption understanding. It will allow for more accurate and contextually nuanced image generation.
Moreover, Imagen 2 addresses common challenges faced by text-to-image tools, such as rendering realistic hands and human faces. The technology incorporates a specialized image aesthetics model. It is based on human preferences, ensuring that generated images align with qualities preferred by users. These advancements collectively contribute to Imagen 2’s improved ability to generate high-quality and realistic images.
The move aligns with Google’s broader strategy of integrating advanced AI technologies across its suite of services.
Google is expanding its generative AI offerings with an upcoming image generator for Google Bard, as revealed in an unreleased changelog dated January 18.
Following the footsteps of tools like the Search Generative Experience (SGE) and Slides, users will be able to create images by entering a few words that describes the image. This new capability is expected to be powered by Google’s Imagen family of models. Imagen is known for their ability to produce high-quality, photorealistic images.
In December, Google introduced Imagen 2, boasting significantly improved image quality. The technology aims at enterprise use cases, such as logo generation or creating images with text overlays.
Google Bard’s image generator is expected to follow safety features similar to those in SGE. The latter, introduced in October, allows users to “draw” or “sketch” prompts directly into the Search field, generating four images at a time.
Just like SGE, Google Bard will also block results with notable people and explicit content
SGE ensures safety by blocking the generation of harmful, misleading, or explicit content. Bard will also restrict prompts featuring notable people or photorealistic faces, reports 9to5Google. Additionally, each generated image will get a metadata labeling and invisible watermarking using Google’s SynthID.
Google Slides and Meet, the latter for custom backgrounds, already incorporate user-facing image generation through Duet AI. The forthcoming addition of an image generator to Bard extends the accessibility of AI-generated content across various Google services.
By adding further descriptions to image captions in Imagen 2’s training dataset, Google is enhancing image-caption understanding. It will allow for more accurate and contextually nuanced image generation.
Moreover, Imagen 2 addresses common challenges faced by text-to-image tools, such as rendering realistic hands and human faces. The technology incorporates a specialized image aesthetics model. It is based on human preferences, ensuring that generated images align with qualities preferred by users. These advancements collectively contribute to Imagen 2’s improved ability to generate high-quality and realistic images.
The move aligns with Google’s broader strategy of integrating advanced AI technologies across its suite of services.
Denial of responsibility! Techno Blender is an automatic aggregator of the all world’s media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, all materials to their authors. If you are the owner of the content and do not want us to publish your materials, please contact us by email – [email protected]. The content will be deleted within 24 hours.