I understand the appeal of teleportation most clearly when I think about the unglamorous parts of travel. The delayed flight, the connection that suddenly becomes a sprint, the small sandwich priced as though it has recently inherited property. Being able to appear beside someone would solve a rather specific collection of problems.
But the stronger pull comes from missing people. A video call helps, sometimes enormously, yet the person still occupies a rectangle. You can see their face and hear the familiar rhythm of their voice while remaining aware that you can't quite share the room.
Microsoft's holoportation research approaches that gap through three-dimensional capture and communication. It creates and sends a changing digital model of a person, allowing a remote viewer to experience that person spatially. The body stays where it is; information travels through the network.
That distinction doesn't make the idea less interesting to me. It gives the wonder a place to stand. We can look at what researchers actually built, what later projects did with it, and why appearing somewhere is still a very different achievement from being there.
What Microsoft actually demonstrated
The clearest starting point is the 2016 Holoportation research paper. It describes a system for reconstructing, compressing and transmitting three-dimensional representations of people in real time. Mixed-reality displays such as HoloLens let a viewer encounter the remote participant within their own surroundings.
Think of the difference between watching somebody turn around on a flat screen and being able to change your own viewpoint around a three-dimensional representation. The second experience gives the viewer a different relationship to the image. The person has a location in a shared spatial arrangement, rather than being confined to a camera's single chosen view.
That's the ambition behind the name. It doesn't describe atoms disappearing at one end and assembling at the other. It describes a communication system intended to make distance feel less visually obvious while leaving the actual participants in separate places.
I like beginning with the research because a demonstration has something concrete to explain. There are cameras, calculations, displays and limitations. The result can be remarkable without requiring us to pretend that a transporter room has quietly opened behind the office kitchen.
A patent isn't a delivery date
Patents are part of this field, but they need to be identified accurately. One relevant Microsoft patent, US10504274B2, concerns fusing, texturing and rendering dynamic three-dimensional models. It describes methods for combining views and handling visual artifacts in a changing model.
That is a useful technical subject, though it doesn't amount to a promise of a freestanding person made of light in every living room. A patent document describes an invention and its claimed scope. It doesn't establish that a particular consumer product is finished, affordable or scheduled for release.
The distinction matters because the phrase Microsoft patented teleportation can make several separate things sound like one announcement. Capture research, rendering methods and display hardware each solve different problems. A filing about one part doesn't automatically settle the others.
I'd much rather keep the specific achievement in view. Combining imperfect camera observations into a convincing moving representation is difficult work. It deserves attention for what it does, without an imaginary shipping date attached to it.
The cameras have to build a person-shaped model
A conventional camera records a view of the scene. A depth camera adds information about distance, helping the system estimate the shape of visible surfaces. Several views can then contribute to a three-dimensional model as the participant moves.
Microsoft's account of its original setup describes eight camera pods, each containing three image sensors. That was a research arrangement designed to capture the subject from multiple directions. It was considerably more equipment than opening a laptop and hoping the webcam understands your best angle.
The system also has to deal with disagreement between views. An arm can hide part of a shirt, a hand can move quickly, and different cameras can see different portions of the same surface. Joining those observations consistently is part of the achievement.
A model that looks reasonable in a still image may reveal problems when somebody walks around it or when the subject moves. The challenge is continuous: capture, update and display have to keep working together. Presence is a demanding customer, and it doesn't politely stop gesturing while the software catches up.
Color has to stay attached to shape
Shape alone wouldn't give us the familiar face we're hoping to see. A system also needs appearance information, including the colors captured by ordinary cameras. That information has to be mapped onto the reconstructed surface in a way that remains coherent over time.
The rendering patent offers a glimpse of the practical difficulties. It discusses blending camera textures and reducing artifacts associated with visibility and imperfect geometry. A camera's useful contribution depends partly on whether it can actually see the relevant surface.
This is where a phrase like turn a person into data starts to conceal the interesting work. The system isn't collecting one perfect, complete description. It's assembling a usable representation from measurements that have limits, while the subject continues behaving like a person rather than a cooperative sculpture.
I find that more engaging than the idea of a flawless digital duplicate. The engineering has to choose what to preserve, how to combine it and how to make the result stable enough for conversation. Every convincing moment rests on a series of less glamorous decisions.
For my own calls about projects, I’d keep a USB webcam in mind for the desk setup. A proper spot to sit, a little light and time to talk through an idea already makes a very useful kind of remote visit.
Then the model has to travel
A changing three-dimensional model creates a substantial communication problem. The system needs to compress its data, send it across a connection and reconstruct a useful view at the other end. Quality and responsiveness have to be considered together.
Microsoft's historical mobile holoportation FAQ describes reducing bandwidth requirements by 97 percent and operating at roughly 30–50 megabits per second for that version. Those figures belong to a particular system and demonstration. They shouldn't be treated as universal requirements for every later form of spatial communication.
The same FAQ discusses the difficulties of lighting changes, vibration and fitting the computing equipment into a vehicle. Taking a system out of a controlled studio creates new problems even when the underlying idea remains the same. Real environments are less considerate than demonstration rooms.
There's something satisfying about those details. A future that sounds ethereal still has to deal with a moving car, a connection and somewhere to put the computers. Even a doorway of light, if we're going to use that metaphor, has a fairly practical installation problem.
Real time still includes delay
Real time is a useful engineering description, but it doesn't mean zero elapsed time. Cameras collect measurements, computers process them, networks carry data and displays produce an image. Each stage can contribute to the delay between a movement and someone else seeing it.
A conversation also needs sound to make sense alongside motion. If a gesture, expression and spoken sentence arrive out of step, the experience can become distracting. A beautiful model isn't the only thing needed for a comfortable exchange.
There isn't one universal threshold that makes every conversation natural or every task safe. The demands of chatting with a friend differ from those of coordinating precise work. A system needs to be evaluated for what people are actually trying to do with it.
I suspect we'd forgive quite a few visual imperfections if the interaction felt easy. We've already held meaningful conversations through grainy video and unreliable audio. But a system claiming to improve presence has to help the conversation, rather than giving everyone another technical issue to discuss before saying hello.
The display is part of the experience
Here's the correction that matters most when imagining the original demonstrations: the remote person wasn't simply standing in open air for every unaided observer to see. Microsoft's research used mixed-reality displays, including HoloLens. Its FAQ also names head-mounted displays as a way to view participants.
A person wearing a suitable display can see a digital object positioned within their view of the environment. Somebody standing beside them without that equipment doesn't automatically see the same thing. The room and the wearer's view of the room are related, but they aren't identical.
The word hologram gets used broadly in technology demonstrations and entertainment. It can refer to experiences built with very different display methods. A life-sized image in a display enclosure, an object rendered through a headset and an optical hologram shouldn't be assumed to have the same mechanism or viewing conditions.
That doesn't mean one experience can't be moving or useful. It means the practical questions come first: what equipment does the viewer need, where can they stand, and what can they actually see? Those details determine the experience far better than the sparkle in a promotional image.
Seeing around somebody can have a purpose
The spatial aspect becomes easier to appreciate when the task involves shape or movement. Imagine explaining a gesture, showing how a sleeve fits or discussing an object from different angles. A fixed camera view can make that awkward, particularly when one participant has to keep repositioning the camera.
A three-dimensional representation could let the remote viewer choose a more useful viewpoint. That is a potential advantage of the format, rather than proof that every interaction becomes better in three dimensions. The benefit depends on the task and the quality of the system.
For an ordinary update about tomorrow's plans, a phone call may already do the job beautifully. Adding capture equipment and a visor wouldn't necessarily improve it. The interesting applications are the ones where spatial information resolves a real difficulty.
I like asking what the extra dimension is doing. If the answer is simply making a meeting look futuristic, the novelty may wear off before the agenda is finished. If it helps somebody understand a movement or communicate a concern, there's a clearer reason to keep developing it.
Telemedicine gives the idea a concrete job
Microsoft's 3D Telemedicine project describes collaboration with reconstructive surgery teams in Scotland and Ghana. The system captures a patient in three dimensions so a remote clinician can inspect the model and discuss care. This is a specific consultation tool, rather than a doctor physically arriving in another country.
The project describes pre- and post-operative assessment, including reconstructive care after burns, cancer or trauma. A remote view of shape and movement may provide information that is difficult to convey through one flat camera angle. The project also incorporates communication tools so patients can participate in the discussion.
An important detail is that this implementation includes a remote viewer application and a monitor for a Teams call. Holoportation's capture and reconstruction techniques don't require every application to reproduce the original headset demonstration. Different viewing arrangements can use related underlying technology.
That makes the story more interesting to me. An idea introduced with the language of teleportation can become a practical way to improve a particular conversation. The result doesn't need to resemble science fiction to matter to the people using it.
Promising research has a defined scope
A 2022 study of the system's development reported three stages: feedback from 23 clinicians, feedback from 26 patients, and a safety and reliability cohort of 40 patients. Patients helped guide the system's development, rather than appearing only at the end to admire the hardware.
The researchers reported improvements over two-dimensional telemedicine in measures including satisfaction, presence and quality. They also examined agreement with face-to-face consultation. These are encouraging findings within the studied setting, with specific participants, measures and clinical tasks.
They don't establish that a three-dimensional consultation can replace every examination or improve every medical outcome. Being able to see a representation doesn't provide touch, take a tissue sample or perform a procedure. The appropriate use still depends on the clinical situation and the evidence for that use.
I find the patient involvement especially worth keeping in the story. A technically impressive system can still be confusing, uncomfortable or impractical. Asking people what to change is part of making an invention useful, even if it doesn't produce the most dramatic headline.
The later story includes open-source software
The project didn't remain frozen in its original demonstration. In a December 2025 account of 3D Telecommunications, Microsoft described releasing the technology under an open-source license for outside researchers and organizations to use and modify. The account also described a mobile installation supporting consultation work in Ghana.
That is a different kind of milestone from announcing a household appliance. Making a codebase available can help other groups experiment and build on previous work. It doesn't remove the need for compatible capture equipment, computing resources, setup and support.
Open source also doesn't make every deployment the same. A team adapting the software for a different task still needs to establish what works reliably in that setting. Medical applications carry questions about evaluation and governance that a successful demonstration alone can't settle.
What appeals to me is the handoff. Research can become useful partly because other people can work with it, question it and adapt it. The future of an idea may look less like one company delivering a perfect finished world and more like many groups solving the next practical problem.
Even the space station has tried it
NASA reported an October 2021 holoportation demonstration aboard the International Space Station. Astronaut Thomas Pesquet spoke with remote participants including flight surgeon Josef Schmid, using a mixed-reality system with software from AEXA. Their live images appeared within his view of the station.
That is a wonderful sentence to be able to write accurately. A doctor on Earth could be represented in three dimensions to somebody in orbit. The demonstration explored communication across a demanding distance without requiring the doctor's body to make the journey.
But space also exposes a limit that no amount of visual realism removes. Signals take time to travel, and communication with Mars can involve substantial delays depending on planetary positions. A more convincing representation doesn't turn a delayed exchange into an instantaneous one.
I still love the possibility of a familiar face appearing in an unfamiliar place. It just needs to live beside the physics and the equipment. Astronauts would need dependable systems, and somebody somewhere would still have to deal with the software behaving strangely at an inconvenient moment.
The product landscape has changed
An older research page can make a project sound permanently current, even when related products have changed. Microsoft's HoloLens release notes identify November 2024 as the final feature release for HoloLens 2, followed by security servicing. That history matters when discussing the hardware associated with the early vision.
Microsoft also says it retired the Mesh web experience, PC and Quest apps, and the Immersive space view in Teams meetings on December 1, 2025. Its product page directs attention toward immersive events in Teams. Those changes should be read as product developments, rather than proof that every underlying research technique has disappeared.
An avatar meeting, a live volumetric capture and a headset are separate pieces of a broad field. One product changing direction doesn't establish that all the pieces have failed, and a new immersive event feature doesn't establish that live photorealistic holoportation has become routine for everyone.
That's why I don't want to promise a date when this will arrive in every home. Research progress and consumer adoption follow different paths. We can recognize real achievements while leaving the eventual form of everyday use open.
Presence is also about the person receiving it
The emotional promise remains powerful. A larger range of gestures and spatial cues could make some distant conversations feel less constrained. But we shouldn't assume that the most realistic representation automatically creates the strongest connection.
Some people may prefer a simple screen, an audio call or an exchange they can answer in their own time. Equipment can introduce discomfort, cost and unfamiliar controls. A technology intended to bring people closer needs to make room for those differences.
There's also the matter of attention. Someone can be physically beside us and only half listening, while a voice on a phone can feel completely present. The quality of a relationship doesn't become a graphics setting just because the image gets better.
I can imagine appreciating a spatial visit and still wanting the ordinary things it can't carry: sharing the same meal, noticing the room's temperature, reaching for someone's hand. We don't have to pretend those absences disappear in order to value the connection that remains.
A likeness needs permission
A detailed recording or live model of a person raises practical questions about consent. Participants should understand what's being captured, whether it's stored, who can see it and whether it can be replayed. These are design questions as well as matters of trust.
A live session and a saved recording are different experiences. Permission to join a conversation doesn't automatically communicate permission for indefinite reuse of a moving likeness. The more personal the representation feels, the more understandable the controls need to be.
Authentication matters too. A convincing appearance isn't proof that the expected person is participating live or that an image hasn't been altered. Any system used for important conversations needs ways to establish identity and communicate what kind of representation is being shown.
None of that requires turning every family call into an interrogation. It asks the technology to make ordinary expectations visible. People should be able to enjoy a conversation without discovering later that their presence became material for an entirely different purpose.
Recorded visits and imagined conversations
The original demonstrations also invited people to imagine replaying a captured visit. A spatial recording could preserve gestures and movement in a way a flat video doesn't. That possibility carries an emotional weight of its own, especially when a recording becomes precious over time.
But replaying a moment is different from receiving a new response. A recording contains what was captured, while an interactive simulation may generate things the person never said. Those distinctions deserve clear labels, particularly when the subject is someone who can no longer clarify their wishes.
I don't think we need to prescribe how everyone should feel about such recordings. Some may find comfort in them; others may find them unsettling or simply unhelpful. A richer format doesn't dictate a single correct way to remember somebody.
The choice should stay with the people involved, including the person whose likeness is captured. A technology that makes an image feel closer should also make it easy to pause, decline or put it away. Wonder works better when it leaves room for an ordinary no.
The doorway still has a frame
What keeps me interested in holoportation is the combination of ambition and stubborn practical detail. Cameras have to see enough, models have to update, connections have to cooperate and the receiving person has to find the result useful. There's alot between a compelling demonstration and a comfortable everyday habit.
I don't need that gap covered with a promise that travel will become obsolete. Sometimes we want to visit a person, and sometimes we want to be in their actual place, with everything unpredictable that comes with it. A spatial call might help with one part of distance while leaving the rest intact.
The real achievement is already unusual enough: a moving representation can be captured in one location and encountered from another viewpoint far away. Researchers have carried that idea into medical consultations, remote demonstrations and experiments in orbit. Those are concrete things to examine and improve.
So I can keep the image of a doorway of light, provided I remember its frame. The frame contains equipment, limits, choices and the person deciding whether to answer. On the other side is someone we may be very glad to see, even when we both know exactly how far apart we still are.
Sources and image note
The linked research and official project pages distinguish demonstrated systems, historical prototypes and later product changes. The three accompanying images are AI-generated editorial illustrations, not Microsoft product photographs or evidence of a headset-free projection system.
Related Reads You Might Enjoy:
We Are Vessels of Light: How Sunlight Moves Through Us and Mends Our Vision
The AI That Writes Its Own Rules: Inside DeepMind’s New Era of Algorithmic Creation
When Robots Run with Humans: Inside the Beijing Half-Marathon for Androids
The Impossible Signal: Mysterious Radio Pulses Beneath Antarctica
Three Months to Mars: How Starship Is Rewriting the Clock on Interplanetary Travel
Through the Shadow of a Giant: What We Learned from Uranus Passing a Star
The Great Attractor: The Mysterious Force Dragging Our Galaxy Toward the Unknown
The Hydrogen Horse: Kawasaki’s Wild Leap into the Future of Movement
