Google's Genie 2 creates playable 3D worlds from text prompts and sample images

Google's Genie 2 enables easy creation of 3D worlds from text and images.

: Google introduces Genie 2, a generative AI model that creates 3D playable worlds from text prompts and sample images. This AI model allows users without prior design skills to rapidly prototype virtual environments. Despite its low-resolution output, Genie 2 can provide various perspectives and complex interactions. Character animation, NPCs, and physics are among its advanced capabilities.

Google has unveiled Genie 2, a new generative AI model designed to simplify the process of creating 3D virtual worlds. By using text prompts and sample images, both user-generated and AI-generated, Genie 2 can produce virtually endless 3D environments that users can control.

The technology provides a consistent albeit low-resolution experience for limited durations, managing various views and complex object interactions. It offers a testing platform for AI researchers and a means for users to experiment with gaming and interactive applications without extensive technical knowledge.

Additionally, Genie 2 supports advanced features such as character animation, NPCs, physics, and lighting. As the AI continues to evolve, Google anticipates improvements that could further enhance its usability for creative and gaming contexts.