Artichoke Labs develops city-scale augmented reality applications by utilizing AI to create precise 3D reconstructions from smartphone video footage, enabling real-time integration of virtual objects into real-world environments. This technology addresses the challenge of accurately mapping and visualizing urban landscapes for enhanced user experiences in augmented reality.
Funding
Funding not disclosed
Founders
Product
Problem
Creating city-scale augmented reality (AR) applications requires precise 3D reconstructions of real-world environments, a process that is traditionally complex and computationally intensive. Existing methods often struggle to achieve the accuracy and real-time performance needed for seamless integration of virtual objects into dynamic urban landscapes.
Solution
Artichoke Labs addresses this challenge by using AI to generate accurate 3D reconstructions of cityscapes from smartphone video footage. Their technology analyzes video frame by frame, identifying key points and transforming them into a spatial representation of the environment. This spatial geometry is then converted into a machine-readable format that is interpreted by a Visual Positioning System (VPS) to locate the position and orientation of the camera in real time, enabling the seamless integration of AR objects into the real world. The AI constantly matches point clouds with the original footage to obtain precise 3D objects even when the camera is moving.
Target Audience
The primary audience includes developers and creators looking to build city-scale augmented reality applications, as well as users interested in creating and sharing AR content on social media.
Features
- AI-powered 3D reconstruction from smartphone video footage
- Real-time Visual Positioning System (VPS) for accurate camera localization
- Point cloud matching for precise 3D object creation
- Spatial mapping for converting geometry into machine-readable format
- Support for integrating virtual objects of any size into real landscapes
- Mobile app (Subview) for creating cinematic AR content