Shutterstock Launches Generative 3D Service, Getty Images Upgrades Image Generation Service Shutterstock and Getty Images have announced the launch of their respective services powered by NVIDIA's Edify AI technology. The announcement was made by NVIDIA CEO Jensen Huang at the Siggraph conference in Denver, Colorado. With the NVIDIA Edify-driven services, Shutterstock and Getty Images enable users to create 3D assets, illuminate virtual environments, or generate images in a shorter amount of time. NVIDIA states that designers and artists now have new and improved methods to enhance their productivity using generative AI based on licensed data. Shutterstock, a creative content platform, has introduced a commercial beta version of its generative 3D service. This service allows creators to quickly produce 3D asset prototypes and generate 360-degree HDRi backgrounds to illuminate scenes using only text or image prompts. Getty Images, a visual content creator and marketplace, has upgraded its Getty Images generative AI service, doubling the speed of image creation, improving output quality, introducing advanced control features, and enabling fine-tuning. These services are built using NVIDIA's Visual AI Factory, which utilizes NVIDIA Edify, a multimodal generative AI architecture. The AI models are then optimized and packaged using NVIDIA NIM, a set of accelerated microservices for AI inference, to achieve maximum performance. Edify enables service providers to train responsible generative models on their licensed data and rapidly scale these models using NVIDIA DGX Cloud, a cloud-first approach that fully leverages the power of NVIDIA AI. Generative AI Accelerates 3D Modeling Shutterstock's service, now available in a commercial beta version for enterprises, allows designers and artists to quickly create 3D objects to aid in prototyping or populating virtual environments. For example, using generative AI, they can swiftly create cutlery and plates on a dining table, allowing them to focus on designing the surrounding characters. The 3D assets generated by this service are ready for editing using digital content creation tools and can be used in various popular file formats. Their clear geometric shapes and layouts provide artists with an advanced starting point to add their own touch. The AI model first provides a preview of a single asset in just 10 seconds. If the user likes it, they can convert the preview into a higher-quality 3D asset, including physically-based rendering materials such as concrete, wood, or leather. NVIDIA states that at this year's Siggraph conference, designers will witness how quickly their ideas can become reality. Shutterstock will showcase a workflow in Blender that allows artists to generate objects directly in their digital viewport. HP will also demonstrate 3D printing and physical prototyping, allowing attendees to design using generative 3D on-site. Furthermore, Shutterstock is collaborating with global marketing and communications services company WPP to bring virtual productions to life using Edify 3D generation technology. Virtual Lighting Becomes Reality Accurately lighting virtual scenes with reflections can be a complex task. Creators often need to manipulate expensive 360-degree camera rigs and create backgrounds from scratch on location or search through vast image libraries for approximations. Now, with Shutterstock's generative 3D service, users can simply describe the precise environment they need using text or images and receive a high dynamic range panoramic image, a 360 HDRi, with an astonishing resolution of up to 16K.