MultiSet SDKs for Visual Positioning

A VPS gives devices precise 6-DoF position and orientation in the real world so AR content and robots can align to physical assets exactly—indoors, outdoors, and at scale. Compared with GPS, Wi-Fi, beacons or QR codes, VPS provides far higher accuracy and persistent spatial awareness, which is essential for navigation, training, inspection, and digital-twin overlays in complex facilities. VPs also provide multiplayer environments which provide a shared sense of space for a fleet of devices.

Integrate MultiSet AI's VPS using our REST API, Unity SDK, or native SDKs for iOS and Android. This ensures seamless integration and precise localization in your AR projects. Our platform also supports on-premises deployments and offline SDKs, keeping your data secure. Designed for various environments, our VPS technology handles everything from small rooms to large stadiums.

MultiSet AI’s neural networks are trained for diverse lighting conditions and dynamic environments. The VPS remains robust and accurate under typical lighting variations, minor physical changes, and the presence of people, ensuring reliable performance. Ideal for manufacturing, warehouse training, and other industrial settings, our VPS technology delivers centimeter-level accuracy even in challenging environments.

A visual positioning system uses camera imagery to calculate a device’s 6-DoF pose. MultiSet’s VPS is scan-agnostic. It fuses LiDAR data, point clouds and 3D scans, delivers sub-centimetre accuracy, and scales to 100,000+ m² venues without markers or beacons.

Any modern phone, tablet, AR headset, robot or drone equipped with a standard RGB camera. Devices with LiDAR (iPhone Pro, iPad Pro, Quest 3) gain extra accuracy but are not required.

Hierarchical indexing holds drift to < 1 cm at ranges under 10 m and < 6 cm at 100 m, with median look-ups of 52 ms on Snapdragon 8 Gen 3 and 38 ms on Apple Silicon.

MapSet stitches many smaller scans into one seamless “mega-map.” It preserves the detail of each fragment while providing a single coordinate system, so users experience uninterrupted navigation across rooms, floors and outdoor areas.

Yes, pass a HintPosition (WGS-84 coordinates from GPS or UWB). MultiSet narrows its search space, localizes up to 40 % faster and reduces on-device CPU/memory load for large and multi environment maps.

Absolutely. Dynamic exposure control adapts to sunlight and shadows, allowing the same map to span factory floors, loading docks and yard space. Yes. Maps can cover large areas and link together for seamless map-to-map transitions (multi-floor, indoor↔outdoor). Independent benchmarking confirms “NavMesh + map stitching” support on device.

Data is encrypted with AES-256 at rest and TLS 1.3 in transit. You can deploy in MultiSet Cloud, a single-tenant VPC, self-hosted Kubernetes, or fully on-device for air-gapped sites - each option meets ISO 27001 and SOC 2 standards.

Unity, native Android, native iOS, WebXR, and Meta Quest SDKs are ready to download, and we’ll build a custom viewport for any camera-based device on request.

MultiSet VPS offers a free sandbox tier for prototyping. Production pricing scales based on active devices, API calls, and map area. Custom SLAs are available for private-cloud and on-device deployments.

Download an SDK, import the sample scene, or book a live demo. Most teams deploy and localize their first device in under 10 minutes.

Export a polygon mesh from your CAD tool (or via a DCC like Blender), apply a PBR base color/diffuse texture, verify real-world scale (1 unit = 1 m), set +Y up, and export to .glb/.gltf. Keep the file lean (ideally <20 MB), with clean UVs and a single material when possible.

Bring your own reality capture: E57 (e.g., Leica), Matterport, NavVis, LiDAR meshes and more. The docs include guided E57 import with meshing improvements for crisp occlusion and smaller files.

Yep. MultiSet supports WebXR endpoints as well as iOS App Clips and Android Instant Apps for “tap-to-try” AR that removes install friction—perfect for pilots and training rollouts.

MultiSet binds content within the map asset for efficient localization. Updates require redeploying the map to devices, which is good for enterprise change control.

Yes—use Unity NavMesh for pathing plus MultiSet’s map stitching to move between areas without re-localizing. This is crucial for multi-floor factories and campuses.

Yes—ROS/robotics integration and HMDs are part of the supported stack, enabling consistent localization across mobile devices, wearables, and robots.

Vision-based systems can fail if a scene changes drastically (e.g., walls removed). Plan maintenance windows to refresh maps and use map stitching for large spaces. (Pro tip: design fallbacks like QR/UWB hints for cold-start.)