World Labs: Build a 3D model of the world from a single image, apply for the Spatial Intelligence model beta test!

General Introduction

World Labs is an AI company focused on spatial intelligence, building Large World Models (LWMs) to perceive, generate and interact with 3D worlds. Founded by world-renowned AI technology pioneer Fei-Fei Li, along with Justin Johnson, Christoph Lassner, and Ben Mildenhall, the team members have outstanding achievements in the fields of computer vision and graphics.World Labs' goal is to drive 3D world reconstruction and new perspective synthesis through advanced AI technologies, providing high-quality spatial intelligence solutions.

Internal test application address:World Labs waitlist

Experience: https://www.worldlabs.ai/blog

Function List

• 3D World Generation: Generate 3D worlds from a single image, providing realistic scene reconstruction.
• Real-time style migration: Real-time style migration techniques applied to Meta, Snap and Prisma.
• Neural Radiation Field (NeRF): Revolutionary 3D scene reconstruction and new perspective compositing methods.
• Pulsar Renderer: Efficient sphere-based micro-renderer with support for real-time NeRFs.
• Space Intelligence Solutions: Scalable real-time neural rendering for virtual reality and Unreal Engine.

Using Help

3D World Generation

1. Upload images: Users can upload single images through the website.
2. Generate 3D models: The system automatically processes the image and generates the corresponding 3D world model.
3. View & Edit: Users can view and edit in the generated 3D model, adjusting the details to meet needs.

Real-time style migration

1. Choose a style: Users can choose their favorite style from a library of preset styles.
2. application style: After uploading an image, the system applies the selected style in real time to generate a stylized image.
3. Download & Share: Users can download the generated stylized images or share them directly to social media platforms.

Neural Radiation Field (NeRF)

1. Upload data: Users can upload multi-view image datasets.
2. Generating NeRF models: The system processes the data and generates a high-quality 3D scene reconstruction model.
3. Interaction and Presentation: Users can interact with the generated NeRF model to explore different perspectives.

Pulsar Renderer

1. Selecting the Rendering Mode: Users can select different rendering modes to suit different application scenarios.
2. real time rendering: The system renders 3D models in real time to provide efficient visualization.
3. Export results: Users can export renderings for further application and presentation.

Space Intelligence Solutions

1. Customized services: Users can contact the World Labs team for customized space intelligence solutions.
2. Technical Support: Provide professional technical support to help users solve problems encountered in the process of use.
3. continuous updating: World Labs regularly updates its technology and features to ensure that users always enjoy the latest AI achievements.