Objects with perception and intelligence. How AI and sensors are shaping the spaces of the future

When the fridge says good morning. Everyday appliances that think ahead, before you ask.

^{Visualization: Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH
Image: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

An exhibit that lifts its head the moment someone enters the room. An armchair that notices someone has just sat down and adjusts the lighting accordingly. Two painted eyes that follow a visitor until they leave again. What sounds like a movie effect has already been built in several real projects, some by hobbyists, some by museums, some by internationally known artists, and increasingly in the furniture industry too.

The technical term for this is Spatial Behavior: how people move through a space, how they react to their surroundings, and how objects in that space can, in turn, react to them. The term originally comes from architecture and behavioral research, not from the tech world.^[1]

This article connects both sides: decades of research into how people perceive space, with today’s sensor and AI technology, which for the first time allows objects, from artworks to furniture, to actually recognize and respond to that behavior, without requiring a multimillion-dollar budget.

• Spatial Behavior describes how people react to their spatial environment.
• The term originally comes from architecture and psychology.
• New sensor and AI technology makes responsive objects affordable today.
• Real examples range from hobby projects to art and furniture design.
• The underlying principles can be applied across many industries.

This article explains where the term Spatial Behavior comes from, what technology is needed for an object to actually recognize and track a person, which real projects already implement this today, from art to the furniture industry, and where else it can be applied.

What Spatial Behavior actually means

The term Spatial Behavior is not a new invention of the tech industry. It has been an officially recognized technical term in psychology and medicine since 1968, and it describes how people and groups react to their immediate surroundings, including the objects within them.^[1]

A central building block of this research is proxemics, coined by anthropologist Edward T. Hall: the study of how people use distance and proximity to one another and to objects in order to communicate.^[2] Anyone who has ever unconsciously taken a step back because someone stood too close has experienced proxemics firsthand, without ever knowing the term.

This exact human behavior, closeness, distance, attention, withdrawal, can now be technically captured and responded to for the first time. A sensor detects how close a person is. Software detects where they are looking. And an object can react to that, the way only a human or an animal could before.

Presence, gaze, distance, response. Spatial behavior made visible.

^{Graphic: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

• Spatial Behavior has been a recognized technical term since 1968.
• Proxemics describes how people use closeness and distance to communicate.
• This behavior was previously only possible between humans or with animals.
• Sensors and software now make it technically detectable for the first time.
• Objects can, as a result, react to people on their own for the first time.

This lays the groundwork for understanding why eye contact in particular, the feeling of being looked at by an object, has become one of the oldest, and at the same time most current, applications of this idea.

When art looks back

Long before affordable sensors and Artificial Intelligence were available, one artist spent three decades exploring exactly this question: what happens when an object looks at a person, instead of the other way around.

Mexican artist Rafael Lozano-Hemmer already showed a giant, painted eye that followed every visitor in the room with “Surface Tension” back in 1992.^[3] It was one of the first interactive artworks of its kind. Since then, he has developed this idea into ever new variations: in “Eye Contact,” hundreds of video recordings of sleeping eyes opened as soon as someone walked by, and in “Zoom Pavilion” (2015), facial recognition software identified entire groups of visitors at once.^[4] In 2024, he presented his most recent version yet with “Binocular Tension”: two large, digital eyes on a display that recognize a viewer, look at them, follow them, and “fall asleep” again once they leave.^[5]

30 years of art that looks back.

^{Infographic: The evolution of Rafael Lozano-Hemmer’s eye-contact works between 1992 and 2024 | Graphic: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

What has changed over these three decades is not the underlying idea, but the effort required to build it. What was an elaborate, one-off gallery commission in 1992 is technically achievable today for far more use cases, as the next chapter shows.

• Rafael Lozano-Hemmer has worked on eye-contact-based art since 1992.
• “Surface Tension” (1992) was one of the first works of its kind.
• “Zoom Pavilion” (2015) used facial recognition for entire groups of visitors.
• “Binocular Tension” (2024) is his most recent version to date.
• The core idea stayed the same, the technical effort dropped significantly.

This long artistic tradition shows: the fascination with objects that look at us is nothing new. What is new is only how accessible the required technology has become.

The technology behind eye contact

For an object to actually recognize, look at, and track a person, it always needs the same six technical building blocks working together, regardless of budget.

First, it needs sensors that monitor the room: a camera or a dedicated motion sensor that detects whether and where a person is located. Second, for the most convincing effect, it needs tiny cameras built directly into the eyes themselves, so the object literally sees, rather than just simulating an externally calculated direction, a technique Disney researchers already presented in 2010 in an animatronic eye that held eye contact and tracked passersby.^[6] Third, it needs software that identifies a person from this image data and tracks their movement, so-called computer vision software.

These three building blocks alone are not enough, however. Fourth, it needs a mechanical implementation: small electric motors, known as servo motors, that actually turn the eyes. The best-known example is the “Animatronic Eye Mechanism” by hobbyist Ikkalebob, a freely available 3D-printing blueprint that has since been used in countless hobby projects.^[7] Fifth, it needs a small computer, a so-called microcontroller, that processes the sensor data in real time and drives the motors accordingly. Sixth, and finally, it needs a behavior model: rules for when the object rests, when it wakes up, how it follows a person, and when it calms down again.

Six building blocks, one moment of eye contact.

^{Infographic: The six core technical building blocks every responsive object needs | Graphic: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

For readers with less technical background, this can be simplified: such an object needs eyes to see, a brain to understand, and muscles to react, just in the form of a camera, software, and a motor.

• Room sensors detect whether and where a person is located.
• Cameras built into the eye itself enable genuine, not simulated, eye contact.
• Computer vision software identifies and tracks the person in software.
• Servo motors mechanically translate the movement into the eyes.
• A microcontroller processes everything in real time, a behavior model controls it.

How differently these six building blocks can be implemented depending on budget and ambition, from hobby project to museum piece, is shown in the direct comparison in the next chapter.

From Halloween hobby to museum piece

The same six technical building blocks can be implemented with very different levels of effort, as a direct comparison of real, independently documented projects shows.

Designer Frederik Van Melle built “The Doorman” in 2018, a 3D-printed Halloween mask that tracks passersby with its eyes, a hobby project whose build instructions are freely available online.^[7] James Brown, a programmer at Weta Workshop, the special-effects studio known for “The Lord of the Rings,” built a face-tracking animatronic skull on his own initiative during a Covid lockdown in 2020. It is now a permanent part of the “Weta Workshop Unleashed” visitor experience in Auckland and mimics visitors’ facial expressions in real time.^[8] At the top end stands Rafael Lozano-Hemmer’s museum-grade “Binocular Tension,” built with custom-developed software and professional 3D sensors.^[5]

Same principle, three completely different budgets.

^{Table: Comparison of real, independently documented projects using eye-contact-based technology | Source: All3DP, Weta Workshop, lozano-hemmer.com | © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

For companies, this range is the actual good news: you don’t have to be a museum to work with this technology, the core building blocks are now accessible for much smaller budgets too.

• The same core technology works from hobby project to museum piece.
• “The Doorman” is a freely accessible, private DIY project.
• Weta Workshop’s skull originated as a personal project during the pandemic.
• “Binocular Tension” is a professional, museum-grade art production.
• The effort scales with ambition, not with the underlying idea.

How this range of technology can be applied to a single object is shown in the following chapter, using the example of a responsive mask.

How a mask could learn to see you

Applied to a concrete object, such as an elaborately designed wall mask, the sequence of the six technical building blocks from chapter 3 becomes easy to follow.

As soon as a person enters the room, sensors or cameras mounted on the outside first roughly detect whether and where someone is present. Small cameras built directly into the mask’s pupils then take over the more precise detection, comparable to the principle Disney researchers already presented in 2010.^[6] Software continuously calculates the person’s exact position in the room from this data. Small servo motors inside the mask turn the eyes accordingly, controlled by a microcontroller that processes all the data in real time. A behavior model finally ensures that the mask rests when no one is there, wakes up as soon as someone is detected, follows the person with its eyes, and calms down again once they leave the room, exactly the behavior Lozano-Hemmer’s “Binocular Tension” also shows.^[5]

From detected person to eye contact in milliseconds.

^{Infographic: Sequence of the six technical building blocks using the example of a responsive wall mask | Graphic: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

Important for context: this specific example is a concept, not an installation that has already been built. It does, however, show how real, independently documented individual technologies can be combined into a coherent overall system.

• Outside sensors first roughly detect whether someone is in the room.
• Pupil cameras take over the more precise detection of the person.
• Software continuously calculates the exact gaze direction from this data.
• Servo motors and a microcontroller translate the movement in real time.
• A behavior model controls resting, waking, following, and calming down.

Why an object that perceives a person this way triggers more than mere technical fascination is explained in the next chapter.

Why this is more than a technical toy

The fact that an object recognizes and looks at a person triggers a reaction that goes far beyond technical fascination, an effect Rafael Lozano-Hemmer himself describes as the core of his work.

Eye contact is one of the most fundamental forms of human communication. When it is triggered by an object, Lozano-Hemmer says, a deliberately “uncanny” experience arises, one that raises the question of who is actually observing whom.^[5] It is exactly this brief moment of unease that captures attention, far more effectively than an object that simply stands there.

For companies, this psychological effect is the genuinely interesting part, independent of the art world. Attention is the scarcest resource in museums, exhibitions, events, and retail spaces. An object that actually perceives a visitor, instead of standing arbitrarily in the room, creates exactly this kind of attention in a way that classic signs, screens, or videos can barely achieve.

The difference between being ignored and being noticed.

^{Graphic: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

• Eye contact is one of the most fundamental forms of human communication.
• When triggered by an object, it deliberately feels unsettling.
• This unease captures attention more effectively than static objects.
• Attention is an especially scarce resource in public spaces.
• Responsive objects achieve this effect differently than classic media.

Where exactly this effect can be applied beyond the art world, including in everyday interior spaces, is shown in the following chapter.

Where this can already be used today

The technology described in this article is not limited to artworks. The same six building blocks can be meaningfully applied across several industries, some already today, others as a logical next step.

In theme parks and entertainment venues, characters could recognize and address individual visitors, instead of showing the same fixed behavior to every guest, a direct evolution of what Weta Workshop already demonstrated as a personal project with its face-tracking skull.^[8] In museums and exhibitions, exhibits could react differently depending on whether a single visitor approaches or an entire group, as Lozano-Hemmer’s “Zoom Pavilion” already did for entire groups of visitors.^[4]

A third field, however, reaches furthest into everyday life: the home and the office. Under the term “Smart Furniture,” a dedicated market for furniture with built-in sensors and connectivity is already emerging.^[9] An early, tangible example was delivered by Toyota subsidiary Aisin as early as 2016, with the design concept “Imagine New Days” by designers Setsu and Shinobu Ito: sofas, beds, and armchairs with built-in weight sensors, of the kind otherwise used in airbag systems, along with LEDs that react to a person’s presence and movement.^[10] How computer vision recognition can be integrated directly into a piece of furniture is also described in a current US patent for a “Smart Furniture Content Interaction System”: a camera built into a chair or sofa detects whether and where a person is sitting, and adjusts content or lighting accordingly.^[11]

Same technology, three different spaces.

^{Infographic: Possible application fields for perceptive objects, from theme parks to museums to living rooms | Graphic: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

For business readers, the decisive point here is this: none of these examples require a completely new technology. All six building blocks from chapter 3 already exist individually and are independently documented, what matters is their deliberate combination for the respective use case, whether an artwork, an exhibit, or a piece of furniture.

• Theme parks could let characters react individually to visitors.
• Museums could let exhibits behave differently depending on visitor numbers.
• “Smart Furniture” is already its own, growing market.
• Aisin showed furniture concepts with sensors and responsive LEDs in 2016.
• A current patent describes camera-based recognition built directly into a chair.

How this principle fits into a bigger, business-relevant picture is shown in the closing chapter.

From novelty to a principle for businesses

Zooming out from the individual examples in art and research, one thing becomes clear: perceptive objects are no longer a rare exception. They have arrived in the present, in three areas that already stand in millions of homes and vehicles today.

In the kitchen, connected refrigerators now greet people in the morning with personalized suggestions, exactly as shown in this article’s opening image. In the living room, lighting and furniture automatically adjust as soon as someone enters. And inside vehicles, sensing technology has long been standard: Mercedes-Benz’s MBUX Interior Assistant recognizes occupants via 3D cameras and independently responds to head, hand, and body movements, “intelligent, reactive, contactless,” as the manufacturer itself puts it.^[13] BMW introduced camera-based gesture control as early as 2015, one of the first systems of its kind in the automotive industry, but since 2023 it has been gradually phased out in favor of voice control and touchscreens, an honest reminder that not every perceptive technology succeeds permanently in the market.^[14]

A digital twin, meaning a digital representation of a real machine, building, or process, is at its core nothing more than a system that captures, understands, and reacts to real-world states.^[15] Exactly the same principle, sensing, recognition, response, is already at work in kitchen appliances, living room solutions, and vehicle interiors today, just applied to a person instead of a machine.

Kitchen, living room, vehicle. Arrived in the present.

^{Graphic: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

For businesses, this means: the technology described here is not an isolated curiosity from the art world, but is already part of everyday life, from the refrigerator to the car seat, just rarely recognized consciously as the same underlying principle.

• Kitchen appliances, living room technology, and vehicles already use this principle.
• Mercedes-Benz’s MBUX Interior Assistant recognizes occupants via 3D cameras.
• BMW’s 2015 gesture control has been gradually phased out since 2023.
• Not every perceptive technology succeeds permanently in the market.
• Same core principle as digital twins: capture, understand, respond.

This brings the article full circle. What began with a painted eye in 1992 has arrived today in kitchens, living rooms, and vehicles, a principle for businesses that want to make spaces, products, or experiences more human-centered. How convincing this principle already is today is shown in the following video.

When a resin skull suddenly talks back to you

The previous chapters explained where the term Spatial Behavior comes from, which six technical building blocks are needed for eye contact, and where this can be applied in practice, from art to your own living room. How convincing this already is today is shown most vividly by one real example.

Embedded here is the official video material of James Brown’s face-tracking animatronic skull at Weta Workshop Unleashed in Auckland. A small camera captures a visitor’s facial expression, software transfers the mouth corners and eye movements onto the servo motors inside the skull in real time, “live puppeteering,” as the developer himself describes it.^[8]

This moment, a static art object suddenly mirroring one’s own facial expression, vividly illustrates what was explained technically in the previous chapters: a counterpart that recognizes, understands, and reacts.

^{Video: Face-tracking animatronic skull, Weta Workshop Unleashed, Auckland | Visuals by Weta Workshop, official project material | Analysis, voiceover script, editing and video production: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

The video makes clear that this technology is not a distant future concept, but has already been a permanent part of a publicly accessible visitor experience since 2020. For businesses thinking about their own interactive installations or products, this example shows how close a smaller version of their own is already within reach.

• Video shows Weta Workshop’s face-tracking skull.
• A camera captures the visitor’s facial expression in real time.
• Software transfers the movement directly onto the servo motors.
• A permanent part of a public visitor experience since 2020.
• A real-world example of how close a custom implementation already is.

This example makes tangible where perceptive objects are heading: from an artistic rarity to a repeatable principle for businesses.

From a perceptive installation to your own Spatial Computing solution

An object that recognizes people and responds to them does not come from a single technology, but from the thoughtful interplay of sensors, Artificial Intelligence, and spatial interaction design, exactly the combination at the core of VISORIC’s work.

The expert team at VISORIC GmbH in Munich combines over 15 years of experience in 3D, AI, and XR with hands-on experience in Spatial Computing, sensor integration, and interactive installations, whether the goal is an exhibit, a brand activation, or a product. VISORIC helps businesses turn an idea into a working, everyday-ready concept.

15 years of experience in 3D, AI, and XR: the VISORIC expert team from Munich.

^{Image: © Ulrich Buckenlei | XR Stager Online Magazine | VISORIC GmbH}

A well-thought-out pilot project, a single exhibit, a single product, can often be realized much faster and more cost-effectively than many businesses expect. VISORIC supports this path from the initial concept idea, through selecting the right sensors and software, to a ready-to-use, permanently deployable solution.

• Advice on sensors and AI recognition for each specific use case.
• Development of interactive, human-responsive installations and products.
• From concept idea to a ready-to-use Spatial Computing solution.

This is exactly where a conversation can start: not with the big, company-wide vision, but with a clearly scoped, quickly implementable first step that shows how perceptive objects can already be used meaningfully today.

Would you like to find out how sensors and Artificial Intelligence could be used for your own installation, exhibit, or product?

Talk to the VISORIC expert team in Munich about Spatial Computing, interactive installations, and modern sensor and AI solutions. Together, we’ll turn your requirements into a precise, everyday-ready concept, with a tangible advantage in attention, impact, and user experience.

Contact:

Email: info@visoric.com
Phone: +49 89 21552678

Sources and references

1. MeSH (National Library of Medicine). Descriptor “Spatial Behavior,” introduced 1968. ncbi.nlm.nih.gov/mesh.
2. YBL Journal of Built Environment. Proxemics of Space, Edward T. Hall. ybljournal.uni-obuda.hu.

3. bitforms gallery / Art21. Rafael Lozano-Hemmer, “Surface Tension” (1992). bitforms.art, art21.org.
4. Studio International. Rafael Lozano-Hemmer, “Zoom Pavilion” (2015) and further works. studiointernational.com.
5. Rafael Lozano-Hemmer. Project description “Binocular Tension” (2024). lozano-hemmer.com.

6. Disney Research. Animatronic eye with eye contact and person tracking, SIGGRAPH 2010. la.disneyresearch.com.
7. All3DP / Instructables (Ikkalebob). “The Doorman,” 3D-printed animatronic mask with sensor and servo eye mechanism. all3dp.com, instructables.com.

8. Weta Workshop. Face-tracking animatronic skull, Weta Workshop Unleashed, Auckland. wetaworkshop.com.

9. Next Move Strategy Consulting. Smart Furniture Market Report. nextmsc.com.
10. Auto Design Magazine. Aisin (Toyota Group), “Imagine New Days,” design concept by Setsu and Shinobu Ito (2016). autodesignmagazine.com.
11. US Patent. “Smart Furniture Content Interaction System and Method,” presence detection via computer vision. image-ppubs.uspto.gov.

12. Mercedes-Benz. MBUX Interior Assistant, official technical description. moba.i.mercedes-benz.com.
13. Tom Bush BMW / TechCrunch. BMW Gesture Control, introduced 2015 and gradually phased out since 2023. tombushbmw.com, techcrunch.com.

15. ISO 23247. Reference framework for digital twins, general definition and core principle.

16. VISORIC case studies in digital twins, real-time 3D, and Spatial Computing.
17. XR Stager platform for real-time 3D, digital twins, Knowledge AI, and industrial Spatial Computing applications.

Contact Persons:
Ulrich Buckenlei (Creative Director)
Mobile: +49 152 53532871
Email: ulrich.buckenlei@visoric.com

Nataliya Daniltseva (Project Manager)
Mobile: +49 176 72805705
Email: nataliya.daniltseva@visoric.com

Address:
VISORIC GmbH
Bayerstraße 13
D-80335 Munich