MultiSet AI

Spatial Intelligence to perceive the real world

[MultiSet AI @multiset_ai]

Speed and Precision with VPS🚀

[Grok @grok]

Current SOTA VPS: MultiSet AI, ranked most robust in 2025 AREA report, offers cm-level accuracy indoors/outdoors, resilient in dynamic lighting.

SOTA visual SLAM: SEGS-SLAM or OGS-SLAM (2025 papers), using Gaussian Splatting for photorealistic mapping and low-drift localization.

Comparison:
- Performance: VPS excels in precision (e.g., <6cm) with pre-maps; SLAM achieves similar (e.g., 2-5cm ATE) but builds maps on-the-fly for unknown areas.
- Efficiency: VPS is faster for queries; SLAM requires more compute for real-time mapping (~17 FPS tracking).
- Reliability: VPS stable in mapped, varied environments; SLAM robust to changes but prone to drift in low-texture or dynamic scenes.

[MultiSet AI @multiset_ai]

This is exactly what our APIs were built for.

Headless spatial. No SDK lock-in. Cloud-offloaded localization. Full control over your camera pipeline.

[suriya @suriyaswamii]

One of the things I’ve been obsessing over recently is how to build a spatial platform with a minimal stack. Building an AR platform currently comes with a lot of bloat, and a headless architecture is something I’ve been exploring.

The key things I wanted to achieve were avoiding platform lock-in and offloading compute to the cloud while still maintaining complete control over the camera pipeline and localization input.

And as always, @multiset_ai had a solution for this. Using their APIs, I built a fully headless spatial layer. And I have made a tutorial so you can build it too!

It was a challenging video to execute, but I think I did a good job! Link below!👇

[MultiSet AI @multiset_ai]

Built using MultiSet VPS SDK for Smart glasses.

[Sadao Tokuyama @tokufxug]

AIメガネのMeta Ray-BanでVPS（カメラで場所を正確に把握する技術）が可能に。

高精度な音声ナビや配車位置特定、複数人での位置共有が可能に。

VPSによる正確な位置の認識により、必要な時と場所で適切なデジタル情報を正確に表示する、新しい空間体験が可能になります。

[Bilawal Sidhu @bilawalsidhu]

Wall hacks in real life using augmented reality.

New mad science experiment -- I used $300 Meta RayBan glasses and an iPhone to recreate EagleEye's "see-through-walls" tracking.

The answer was visual positioning tech to build a shared spatial map. Code on GitHub.

0:00 - The Experiment
0:38 - Dumb Glasses, Smart Tracking
1:54 - No Street View? No Problem
3:17 - Seeing Through Walls With a Phone
4:24 - The Indoor Problem Nobody Solved
5:20 - This Is What It Can Do
6:38 - Use Cases Beyond the Flex
7:44 - The Double Edge
10:32 - Conclusion

[Ty @0to60in2point9]

@bilawalsidhu has one of the best @YouTube channels. Top tier, cutting edge application of tech. @multiset_ai https://t.co/ezx343JJPU

[Nikhil Sawlani @sawlaninik]

It was great hosting a super talented group @multiset_ai Bangalore office

[MultiSet AI @multiset_ai]

Join us in Bengaluru to talk about VPS use cases in XR, AI and robotics.

[suriya @suriyaswamii]

On the sidelines of @ycombinator's Startup School India. We are teaming up with @multiset_ai and hosting an XR Meetup for folks in Bengaluru!

If you are someone who is building in this space or is interested in AR and VR. Make sure you join! We have some really great people joining!

We have @sawlaninik joining, who has built some truly great products, including @Multiset and ARWay.

@Virendra0698 will talk about @anomalysims and all his other truly great projects. I bet you have seen him a lot on your timeline!

And @RangeshUs will be talking about how he is building the next gen of apps using Smartglasses.

Make sure you RSVP using the Luma below. See you soon!

[MultiSet AI @multiset_ai]

Meta just featured MultiSet AI on their official developer blog.

We built on the Wearables Device Access Toolkit to stream camera frames from Ray-Ban Meta glasses directly into our VPS engine. Real-time location tracking. Navigation. Hands-free.

[Meta for Developers @MetaforDevs]

The Wearables Device Access Toolkit is in developer preview.

Extend your app with camera and audio inputs and build real-time, hands-free experiences.

Early integrations are already live—check them out. https://t.co/QOdj8ooXmq

[MultiSet AI @multiset_ai]

Check out the new On-Device VPS example by @suriyaswamii

[suriya @suriyaswamii]

With this awesome feature, @multiset_ai also allows you to run your AR application fully offline, on-device. You can download your map once or embed it in your APK itself. And the way it works is pretty neat! If you are looking to build on top of this, check my tutorial linked below👇

[MultiSet AI @multiset_ai]

https://t.co/m5d0HKwA4d

[MultiSet AI @multiset_ai]

Why have users unwittingly helped build fragmented 3D map of the world, isolated hotspot scans of public spaces with little meaningful connectivity 🌍, when you can run the world’s most robust, private VPS in your own cloud ☁️🔒?

[MultiSet AI @multiset_ai]

Amazing example of VPS 6DoF Tracking on smart glasses. Kudos to @BrilliantSole

[Brilliant Wear @BrilliantSole]

VPS tracking on smart glasses

Experimenting with @multiset_ai’s Visual Positioning System (VPS) to determine a person’s position relative to a 3d scanned environment, using just an image

Works with compatible smartglasses, in this case @omidotme’s Omi Glass (based on @seeedstudio’s ESP32-S3 Sense) running our custom firmware

[MultiSet AI @multiset_ai]

6DoF Object Tracking &lt;&gt; Cross Platform &lt;&gt; Device Agnostic.
Thank you for sharing @suriyaswamii

[suriya @suriyaswamii]

Tutorial video for Object Tracking with @multiset_ai is up on YouTube now. link below👇

It's really interesting how you can scan any object with polycam etc. And turn that GLB into a fully tracked object anchor in your AR scene.

[MultiSet AI @multiset_ai]

Check out the MultiSet VPS sample for Meta Ray-Ban Glasses: https://t.co/8OZ0Zd58Dp

[Oscar Falmer @OscarFalmer]

📍 VPS (Visual Positioning System) by @sawlaninik from @multiset_ai

A sample iOS application demonstrating Visual Positioning System (VPS) integration with Meta Ray-Ban Smart Glasses. This app showcases localization and turn-by-turn navigation using the MultiSet VPS API and Meta Wearables Device Access Toolkit.

> https://t.co/6IpJ49Q7G7

[Nikhil Sawlani @sawlaninik]

Smart glasses just got spatially intelligent. 🚀

@multiset_ai VPS now supports wearable devices, starting with Meta Ray-Ban smart glasses. With &lt;5 cm location accuracy, glasses gain a precise understanding of device position.

[Bilawal Sidhu @bilawalsidhu]

Made using @multiset_ai and on device visual localization: https://t.co/WPLnoQkkn9

[MultiSet AI @multiset_ai]

Real world AI for AR!!

[田中義弘 | taziku CEO / AI × Creative @taziku_co]

派手なAR表現より、実はこっちの方が重要。

MultiSet AIのVisual Positioningは、暗所・リアルタイム・端末内処理で自己位置推定が成立している。

ロボットもARも「自分がどこにいるか」を理解できなければ何も始まらない。この基盤技術、もっと評価されるべきだと思う。

[Bilawal Sidhu @bilawalsidhu]

A great visual positioning system makes augmented reality feel like magic... this tech helps your phone (or robot) figure out precisely where it is in 3D space.

Here's MultiSet AI nailing it in real-time, at night, and on-device too. This new reveal shader animation shows just how well it's working.

IMO this tech is slept on despite being key infrastructure for physical AI to actually work.

[MultiSet AI @multiset_ai]

On-Device large-scale localization is key infrastructure needed for Physical AI

[Bilawal Sidhu @bilawalsidhu]

A great visual positioning system makes augmented reality feel like magic... this tech helps your phone (or robot) figure out precisely where it is in 3D space.

Here's MultiSet AI nailing it in real-time, at night, and on-device too. This new reveal shader animation shows just how well it's working.

IMO this tech is slept on despite being key infrastructure for physical AI to actually work.

[MultiSet AI @multiset_ai]

@HoloLabInc Thank you for sharing!

[ホロラボ @HoloLabInc]

ここ最近、調査している MultiSet( @multiset_ai )というサービスにある VPS を試してみました。記事です。
LeicaやFaroの点群のほかに、最近アツイXGRIDSの点群も利用できることが特長で、2025年9月1日に一部の新規販売が終了したVuforiaの代替として期待です。
https://t.co/KCSbiDVKQo