Tech Desk 
Dec 04, 2024
Google Vertex artificial intelligence (AI) platform officially launched two latest AI models including Imagen 3 and Veo. These models empower users to create and animate visual content with great precision, top-notch features, and more.
Imagen 3: Enhanced image generation
Imagen 3 represents a significant leap from its predecessors, excelling in detail, lighting, and artifact reduction. It allows users to generate images using text prompts for those on the “allowlist,” including:
- Inpainting: Modify existing images by adding elements.
- Outpainting: Expand an image beyond its original borders.
- Background replacement: rIt automatically replace image backgrounds effortlessly.
Businesses will provide reference images to use along with the prompt. According to the Alphabet-owned Google, the Imagen 3 customisations enable users to “infuse your own brand, style, logo, subject or product features when generating new images.”
Imagen 3 will be accessible to all Vertex AI users starting next week.
Veo
Veo, available in private preview, focuses on animation and video generation.
Here are the top-notch features, offering an improved user experience.
- Animating static images: Users can provide an image along with directives to animate them.
- Text-to-video creation: Generate videos entirely from the ground up by using only a text prompt, providing unparalleled flexibility.
Both Imagen 3 and Veo integrate DeepMind’s SynthID technology, ensuring AI-generated visuals are invisibly watermarked to prevent misuse or misrepresentation as real content.
Moreover, Google reiterated its commitment to ethical AI development, highlighting that customer data was not used while training any of these models.
