Google Cloud announces Imagen 2 text-to-image generator


Google Cloud has launched Imagen 2, the newest improve to its text-to-image capabilities.

Obtainable for Vertex AI clients on the allowlist, Imagen 2 permits customers to craft and deploy photorealistic photographs utilizing intuitive tooling and fully-managed infrastructure. 

Developed with Google DeepMind know-how, Imagen 2 affords improved picture high quality and a variety of functionalities tailor-made for particular use instances.

Key options of Imagen 2 embrace:

  • Numerous picture era: Imagen 2 excels in creating high-resolution photographs from pure language prompts that cater to numerous consumer necessities.
  • Textual content rendering in a number of languages: Overcoming frequent challenges, Imagen 2 helps correct textual content rendering in a number of languages.
  • Brand era: Companies can leverage Imagen 2 to create quite a lot of inventive and practical logos—with the choice to overlay them on merchandise, clothes, enterprise playing cards, and extra.
  • Captions and question-answering: Imagen 2’s superior picture understanding capabilities facilitate the creation of descriptive captions and supply detailed solutions to questions on picture components.
  • Multi-language help: Imagen 2 introduces help for six further languages in preview, with plans for extra in early 2024. This consists of the power to translate between immediate and output.
  • Security measures: Imagen 2 incorporates built-in security precautions, aligning with Google’s Accountable AI rules. It options security filters and integrates with a digital watermarking service to make sure accountable use.

Enterprise-ready capabilities

Imagen 2 on Vertex AI is designed to fulfill enterprise requirements, providing reliability and governance akin to its predecessor. With new options comparable to high-quality picture rendering, improved textual content rendering, brand era, and security measures, Imagen 2 goals to supply organisations with a complete instrument for inventive picture era.

Main corporations like Snap, Shutterstock, and Canva have already embraced Imagen for inventive functions.

Chris Loy, Director of AI Providers at Shutterstock, commented: “We exist to empower the world to inform their tales by bridging the hole between thought and execution.

“Selection is important for the inventive course of, which is why we proceed to combine the newest and biggest know-how into our picture generator and modifying options—so long as it’s constructed on responsibly sourced knowledge,”

Danny Wu, Head of AI at Canva, added: “We’re persevering with to make use of generative AI to innovate the design course of and increase creativeness.

“With Imagen, our 170M+ month-to-month customers can profit from the picture high quality enhancements to uplevel their content material creation at scale.”

As Imagen 2 makes waves within the inventive trade, organisations are inspired to discover its potential. Google Cloud anticipates customers will harness the brand new options to raise their inventive endeavours and construct on the success achieved with Imagen.

(Photograph by G on Unsplash)

See additionally: Microsoft unveils 2.7B parameter language model Phi-2

Need to study extra about AI and massive knowledge from trade leaders? Try AI & Big Data Expo happening in Amsterdam, California, and London. The excellent occasion is co-located with Digital Transformation Week.

Discover different upcoming enterprise know-how occasions and webinars powered by TechForge here.

Tags: ai, artificial intelligence, deepmind, enterprise, Google, google cloud, imagen 2, text-to-image, vertex ai



Source link

Exit mobile version