Google Veo and Imagen 3 lead AI innovation, image generation and re-evolution!

Generative AI is driving a wave of digital transformation that is driving businesses to accelerate growth. According to a recent study, 86% of companies that have introduced generative AI report increased revenue, with an average revenue increase of up to 6%. On behalf of generative AI can lead a strong market momentum, Google has also continued to improve technology in the field of AI, and has launched the industry-leading image generation model Veo and image generation model Imagen 3. To meet the growing demand for AI technology, Google Cloud announced that the two models will be integrated into the Vertex AI platform to provide more convenient and comprehensive AI services to help enterprises quickly develop innovative applications.

1. Veo is now available in a private preview on Vertex AI, making it easy to generate high-quality videos based on simple text or image prompts. As the first hyperscale vendor to offer image-to-image models, Google is helping businesses turn existing creative assets into dynamic visuals. This breakthrough technology opens up new possibilities for creative expression and streamlines the workflow of film production.
2. Imagen 3 will begin to be available to all Vertex AI customers. Imagen 3 surpasses earlier versions of Imagen in detail, lighting, and artifact reduction by producing the most realistic, highest quality images with simple text prompts. Businesses can quickly generate high-quality images and use them for marketing, advertising or product design.

Vertex AI provides an orchestration platform that makes it easy to customize, evaluate performance, and deploy models on infrastructure. As for the privacy section, don’t worry at all! In line with Google Cloud’s AI principles, Veo and Imagen 3 are developed and deployed on Vertex AI with built-in precautions such as digital watermarking, security filters, and data governance.

Veo’s most powerful movie generation model is now available on Vertex AI

Developed by Google DeepMind, Veo can generate various videos and visual styles based on text or image cues at high speed and in high quality. With a deep understanding of language and vision, Veo can produce films that are tightly integrated with cues. Veo on Vertex AI produces consistent and coherent shots in which people, animals, and objects are realistic. See the following example of Veo image-to-image generation on Vertex AI: 

Image to image: Veo generates videos from existing images or flat photos generated by AI. The following example shows how Veo can create a short movie clip using images produced by Imagen 3 (the top two images) and real-world images (the bottom two images).

Text to Video: Here is an example of how Veo uses text to create short videos.

 Veo on Vertex AI enables businesses to easily generate high-quality videos based on simple text or image prompts. This means faster production, lower costs, and the ability to produce video content quickly. Veo’s technology can be an important partner in human creativity, allowing creators to focus on higher-level tasks, while AI can help handle tedious or repetitive aspects of filmmaking. Clients such as Agoda are leveraging the power of AI models such as Veo, Gemini, and Imagen to simplify video AD production and significantly reduce production time. Whether you’re a community editor creating creative posts, a sales team creating compelling proposals, or a creative team exploring new concepts, Veo can streamline your workflow and open up new visual storytelling possibilities.

Imagen 3: Our highest quality image generation model, now fully available on Vertex AI

Imagen 3 is the highest quality image generation model in Google Cloud, requiring only text prompts to produce an incredible level of detail to produce realistic, lifelike images.

All Google Cloud customers will be able to access Imagen 3 on Vertex AI. With Imagen 3 on Vertex AI, you can generate high-definition images and movies with simple text prompts. See the following example of Imagen 3 image generation capabilities:

Imagen 3 editing provides a powerful and user-friendly way to refine and customize any image. You can edit photos with simple text prompts, edit only part of the image, including updating the product background, or enlarge the image to meet size requirements.

The Imagen 3 Customization allows you to better generate images of the desired features through the image samples provided. Brands can also use generative AI to create images that match your brand, product, and style, opening up new creative possibilities that shorten the marketing process and accelerate the development of ideas.

Under generative AI, corporate privacy can still be protected

Secure and privacy-sensitive functionality is critical for AI models. According to Google Cloud’s AI principles, Veo and Imagen 3 on Vertex AI are completely free from privacy concerns, and their models are built with security at their core.

• Digital watermarking: Google DeepMind’s SynthID imparts an invisible watermarking into every image and image generated by Imagen 3 and Veo, helping to reduce error messages and misattribution issues.
• Security Filters: Both Veo and Imagen 3 have built-in protections that help prevent the creation of harmful content and adhere to Google’s AI principles. We will continue to invest in new technologies to improve the security and privacy protection of the models.
• Data Governance: According to Google Cloud’s built-in data governance and privacy controls, customer data is not used to train the model, and your customer data is only processed according to your instructions.
• Copyright Compensation: Google’s compensation for generative AI services gives you peace of mind by addressing copyright issues in an industry-first way.

Customers create value through Veo and Imagen on Vertex AI

Leading consumer goods company Mondelez International (which includes Chips Ahoy!) Brands such as Cadbury, Oreo and Milka are using generative AI to accelerate and enhance campaign content creation, with more than 100 brands rapidly producing localized visual material for countries.

Agoda is a global digital travel platform that offers great deals on more than 4.5 million hotels, flights and activities, allowing travelers to easily travel the world at low prices. Agoda is now testing Imagen and Veo on Vertex AI to create visuals that will allow the Agoda team to generate unique images of travel destinations that can then be used to generate videos.

An example of how Agoda’s marketing team uses AI models such as Veo and Imagen to help create promotional videos.

“At Agoda, we’re committed to helping people see the world for less and make travel experiences more convenient. We’re exploring Google Cloud AI’s media generation capabilities, using Imagen to create unique visuals of dream destinations in a variety of styles. These then turn them into video through Veo’s experiments with image-to-video technologies that have the potential to shorten our content creation process from days to hours. – Matteo Frigerio, Agoda’s Chief Marketing Officer

This article has been translated and adapted from the official Google Cloud blog to provide you with the most cutting-edge technology trends and application examples. This article explores how Google Cloud’s new Veo and Imagen 3 can quickly generate accurate images and images on the Vertex AI platform, enabling enterprises to achieve more efficient and accurate applications. As an elite partner of Google Cloud, Microfusion will continue to promote the practical application of these innovative technologies and help enterprise customers seize the business opportunities brought by AI. If you’re interested in learning more about how Vertex AI can accelerate your work, contact us, and we look forward to being your best partner for digital transformation. If you are interested in the multiple applications of Google Cloud, please pay close attention to our event information and look forward to seeing you at the event!