Get the newsletter
Meta · AR / VR · Filed Dec 18, 2024 · USPTO published Jun 18, 2026 · Jun 18 release

Meta Patents Technology That Stops Virtual Hands Passing Through AR Buttons

Anyone who has used a VR or AR headset knows the weird moment when your virtual hand phases right through the button you're trying to press. Meta's latest patent is a direct attempt to fix that.

By · Updated · 3 min read

Meta Patent: Intent-Based Hand Tracking in AR Headsets, figure from US 2026/0169549 A1
Figure from the official USPTO publication.
Publication number
US 2026/0169549 A1(US20260169549A1)
Applicant
Meta Platforms Technologies, LLC
Filed · Published
Dec 18, 2024 · Jun 18, 2026
Inventors
Jennifer Lynn SPURLOCK, Norah Riley SMITH, Andrew C. JOHNSON, Matthew Alan INSLEY, Stella MUEHLHAUS, Brandon FURTWANGLER
Status when we published
Approved; patent expected soon
Document
20 claims
US classification
345/156
Examiner
GIESY, ADAM (Art Unit 2622)

What Meta's virtual hand 'locking' actually does

Imagine you're wearing a mixed-reality headset and you reach out to tap a virtual button floating in front of you. Right now, the system just mirrors your real hand as a digital copy — so if your real finger pushes past where the button should be, the virtual hand awkwardly passes straight through it. It looks broken and feels wrong.

Meta's patent describes a smarter handoff. When the system detects that you intend to press something — based on your hand's direction or how close it is — it breaks the one-to-one mirror. Your virtual hand stops chasing your real hand and instead locks onto the virtual button, compressing it the way a real button would compress. Meanwhile, a separate, ghost-like image of your actual hand continues to show where your fingers really are.

The result is that the button behaves like a physical object — it squishes, stops at the bottom, and doesn't let your hand clip through — while you can still see what your real fingers are doing. It also quietly limits how much wrist rotation the virtual hand shows, so the whole interaction looks natural and grounded rather than rubbery.

How the system detects intent and splits hand rendering

The patent describes a three-part system that kicks in the moment the AR headset decides the user is about to interact with a virtual object.

Step one — intent detection: The headset watches for either a directed gesture toward a virtual object or the physical hand coming within a threshold proximity of it. Crucially, the specific intent the first independent claim focuses on is a pressing motion — the system recognizes when the user is pushing toward a virtual button or panel.

Step two — compression and locking: As the press begins, the virtual hand is rendered as if it's physically compressing the object (think of pressing a soft key on a touchscreen). Once the virtual object hits its maximum compression — a simulated bottom-out point — the system decouples the virtual hand from the real hand's position entirely. The virtual hand stays locked to the compressed object even if the real hand pushes further past that point.

Step three — dual-layer rendering: With the virtual hand locked, the system simultaneously renders a second, separate representation of the real hand that continues tracking the user's actual fingers. The patent also describes limiting wrist-rotation display on the locked virtual hand, preventing visually jarring contortions while the hands are in disagreement about position.

What this means for Quest and future AR headsets

For anyone using Meta's Quest headsets or future AR glasses, this kind of detail is the difference between an interface that feels like a toy and one that feels like a real tool. The classic complaint about virtual interfaces is that touching things feels hollow — your hand passes through objects with no sense of physical consequence. Locking the virtual hand to a compressed button gives the interaction a tactile logic even without actual haptic feedback.

This also signals where Meta is investing engineering effort: not just in making the virtual world look good, but in making the interaction model feel physically credible. As Meta pushes mixed-reality use cases — productivity, communication, spatial computing — getting button and touch interactions right is a foundational problem worth solving properly.

Source: US 2026/0169549 A1 at the USPTO (PDF). Full patent text and figures are in the official publication.

Get our take in your Top Stories

Liked this breakdown? Add Patentlyze as a preferred source on Google, and our plain-English take shows up more often in your Top Stories the next time Meta patent news breaks.

Add Patentlyze on Google One click.
There are more where this came from

We read every patent application Big Tech publishes and send you the ones worth knowing. Plain English, free, every week.

More from Meta

All 190 Meta patents →

More in AR / VR

All 690 AR / VR patents →