Smart Eye, the Swedish firm whose software already watches drivers in dozens of production cars, said on June 9 that the cabin camera can now estimate a driver’s heart rate and breathing rate. It works with no wearable and no physical contact. The same lens that already tracks where the eyes point reads the pulse in the face.
The company will demonstrate the feature publicly for the first time this week at InCabin USA in Detroit, the in-cabin sensing conference running June 9 through 11 at Huntington Place.
The capability is newer as a product than as a science project. Smart Eye first showed camera-based heart-rate and respiration tracking at InCabin Brussels in June 2023.
The 2026 announcement moves it toward something automakers could order, and bundles it with the alcohol-impairment detection the company brought to CES this past January.
The pitch has changed. It is no longer “look what a camera can do,” but “look what a camera you already have to install can do.”
That second sentence is the real story.
Key Takeaways
- No new sensor. The heart-rate and breathing estimates run on the same camera-based driver monitoring system (DMS) hardware automakers are already fitting. Smart Eye’s whole argument is the missing line on the bill of materials: nothing extra to add.
- The technique is rPPG. Remote photoplethysmography reads tiny color shifts in facial skin as blood pulses through it. Smart Eye uses near-infrared illumination so it works in the dark, then fights to pull that faint signal out of head movement, cabin vibration, shifting light, and a face partly blocked by a hand or sunglasses.
- Regulation is the engine. The European Union already mandates driver drowsiness and distraction warning on new cars, which is putting cabin cameras into the fleet by law. Once the camera is there, vital signs are a software upgrade, not a redesign.
- Cameras are not the only contender. Tesla switched on cabin radar in early 2025 that detects a heartbeat through a blanket to find a child left behind. Radar handles darkness and occlusion better; cameras win on reusing hardware the car needs anyway. That split is the live engineering debate.
- It is not in a car you can buy yet. Smart Eye named no automaker, no model, no launch date, and published no accuracy figures. A trade-show demonstration is a long way from a safety function that decides to slow a car down.
- Heart rate is the easy half. Smart Eye’s own wording is telling. Heart rate is “detected.” Breathing rate is “estimated.” The gap between those two verbs is the gap in the underlying confidence.
The News Underneath the Headline
A camera reading your pulse sounds like the headline. The headline is the cost structure.
Driver monitoring is no longer an option box in Europe.
Under the EU’s General Safety Regulation, Driver Drowsiness and Attention Warning has been required on newly type-approved cars since July 2022, and on every new registration since July 2024. The tougher Advanced Driver Distraction Warning rule reaches every new registration in July 2026, the same month Smart Eye picked for this announcement.
Both rules push a driver-facing sensor into the cabin. In most designs, that sensor is a camera.
Once a car carries a camera aimed at the driver’s face by law, asking it to also estimate a pulse costs only software. The feature adds no part and changes no wiring.
That is why a heart-rate feature first shown three years ago is suddenly worth a press release. The hardware finally lives in the cars.
Martin Krantz, Smart Eye’s chief executive and founder, framed it as a progression. “Vehicles are becoming better at understanding the driver’s condition in real time, starting with distraction and drowsiness and now expanding into areas like impairment detection and vital signs monitoring,” he said.
The safety case he offered is the obvious one: catch a medical emergency early. “If warning signs of sudden illness can be detected early enough, it could eventually help reduce accidents, improve response times, and support better outcomes in post-crash scenarios.”
Those hedges, “could” and “eventually,” do a lot of work in that sentence. Keep them in mind through the rest of this.
How a Camera Reads a Pulse
The method has a clinical name, remote photoplethysmography, and a simple physical basis. Every time the heart beats, blood floods the capillaries just under the skin. That extra blood absorbs slightly more light, so the skin darkens by an amount no human eye can register.
A camera can.
Point it at a face, watch a patch of forehead or cheek across many frames, and the rhythmic flicker in reflected light traces the heartbeat. The same waveform, modulated more subtly, carries clues to breathing.
Doing this on a couch in a lab is one problem. Doing it in a moving car is another.
The signal is faint to begin with, and a driving cabin buries it in noise. The head turns and nods. The body jostles over expansion joints. Sunlight strobes through roadside trees, oncoming headlights wash across the face at night, and a hand or a coffee cup blocks part of the view.
Smart Eye lists those exact four enemies, head movement, cabin vibration, changing light, and facial occlusion, as the things its algorithms separate from the real pulse in real time.
How well it survives that gauntlet is the question Smart Eye did not answer. Independent research says the result depends heavily on conditions.
The most cited in-vehicle study, the SparsePPG work presented at a 2018 computer-vision conference, measured camera heart-rate error against a reference monitor. It found the error roughly doubled inside a small, vibrating car compared with a relaxed passenger: a mean near 3.4 beats per minute for a seated passenger, closer to 7.9 in a compact car.
That study used different hardware and predates Smart Eye’s current system by years, so treat it as a floor for context, not a verdict on the new product.
But it sets a fair expectation. A number on a screen is not the same as a number you would trust to call an ambulance.
Cameras Versus Radar
Smart Eye is not racing alone, and its sharpest competition does not use a camera at all.
Radar reads vital signs too, and it shipped first.
Tesla switched on a 60-gigahertz cabin radar through a software update in February 2025 that senses a heartbeat and breathing well enough to find a child or pet left in a parked car, even under a blanket or inside a rear-facing seat.
Vayyar sells in-cabin imaging radar built around the same idea. Cipia, an Israeli monitoring specialist that Samsung’s Harman unit bought in June 2025, showed a system pairing a 60-gigahertz radar with an infrared camera at CES.
Radar’s advantages are real. It shrugs off darkness, sunglasses, and a hand over the face, and it reaches passengers a driver-facing camera cannot see.
The camera’s advantage is the one Smart Eye keeps pointing at. Radar is a part you have to add and pay for. The DMS camera is a part the car already needs to clear regulation.
For an automaker counting pennies across millions of units, “free” beats “better” more often than engineers like to admit.
Seeing Machines, the Australian firm that is Smart Eye’s closest rival in camera-based monitoring, has built its business on the same regulatory tailwind. It has not put a shipping vital-signs product on the table the way Smart Eye now has.
What Would Have to Be True
The safety story is appealing. A driver suffering a heart attack or a stroke usually gives warning signs in the seconds before losing control, and a car that noticed could brake, pull over, or summon help.
NHTSA’s crash-causation survey attributes roughly 1.3 percent of crashes to a driver’s medical episode, with blackouts, seizures, and diabetic events leading the list. That is a small slice of a large number, and the worst of those crashes are the kind a watchful cabin might soften.
Turning the appeal into a feature takes three things Smart Eye has not shown: a named automaker willing to ship it, published accuracy from inside a moving vehicle rather than a conference booth, and a defined response. A heart-rate readout the car does nothing with is a gauge, not a safety system.
Europe’s Euro NCAP roadmap points at where this is meant to land. From 2026, its crash-test scoring adds a post-crash phase that imagines in-cabin sensors reporting a victim’s heart rate and breathing to emergency responders.
The regulatory destination exists. The production hardware does not, yet.
Who Smart Eye Is
The company is not a startup chasing a headline.
Founded in Gothenburg in 1999, Smart Eye spent two decades in eye-tracking for research labs and aerospace before automotive became its center of gravity. It bought the emotion-AI firm Affectiva and the biometric-research platform iMotions in 2021, and it now lists customers from Volvo, BMW, GM, Polestar, and Geely to NASA and Boeing.
By the close of 2024, it counted 357 automotive design wins across 22 automakers, with software live in 75 car models. Revenue reached 355 million Swedish kronor that year at a 90 percent gross margin, though the company still ran an operating loss as it scaled.
That track record is why this announcement deserves more than a shrug. It is also why the missing details stand out.
A firm with software in 75 production models knows how to announce a real design win when it has one. This was not that.
It was a capability demonstration, timed to a Detroit trade show and a July regulatory deadline, from a company that wants automakers to know the feature is ready when they are.
Real Capability, Demo-Stage Product
Smart Eye has shown that the camera already going into your next car can take a credible guess at your heart rate, and a rougher one at your breathing, without any new hardware. That reuse, not the biometrics, is what makes it likely to reach production: regulation is installing the cameras, and vital signs ride along as software. The caution is equal to the promise. There is no named automaker, no launch date, no accuracy data from a moving car, and no described action the vehicle would take. Radar rivals already detect a heartbeat in shipping cars, and the published research says a camera’s pulse reading degrades exactly where you need it most, in motion. The capability is real. The product is still a demonstration. Watch for the first automaker to put a date on it.
Frequently Asked Questions
What did Smart Eye announce?
On June 9, 2026, Smart Eye said its driver monitoring software can now estimate a driver’s heart rate and breathing rate using the existing cabin camera, with no wearable device or physical contact. The company is demonstrating it for the first time at InCabin USA in Detroit, June 9 to 11. It named no automaker, model, launch date, or accuracy figure.
How can a camera measure my heart rate?
Through remote photoplethysmography, or rPPG. Each heartbeat pushes blood into the capillaries under your facial skin, which changes how the skin reflects light by an amount too small to see but large enough for a camera to track across many frames. Software reads that rhythmic flicker and converts it to beats per minute. Breathing is derived from subtler modulations in the same signal.
Do I need to wear anything?
No. That is the central claim. The estimate is fully remote and contactless, using only the camera already pointed at the driver. No watch, strap, or clip is involved.
Can I buy a car with this today?
Not as a heart-rate feature from Smart Eye. The company announced a capability and a demonstration, not a production design win, and disclosed no automaker or date. Separately, some cars already use a different sensor for related jobs: Tesla activated cabin radar in 2025 that detects a heartbeat to find a child left in a parked vehicle.
How accurate is camera-based heart rate in a moving car?
Smart Eye did not publish a figure. Independent research offers context: a 2018 in-vehicle study measured camera heart-rate error near 3.4 beats per minute for a seated passenger and closer to 7.9 inside a compact car, meaning accuracy roughly halves once motion and vibration enter. That used older hardware, so treat it as a baseline rather than a measure of Smart Eye’s current system. The honest summary is that camera pulse reading is hardest in exactly the conditions a moving car creates.
Why are cars getting driver-facing cameras at all?
European law. The EU General Safety Regulation requires Driver Drowsiness and Attention Warning on all new registrations since July 2024, and Advanced Driver Distraction Warning on all new registrations from July 2026. Most automakers meet those rules with a cabin camera. Once that camera is installed, additional capabilities like vital signs become a software question rather than a hardware one.
Camera or radar, which is better for vital signs?
They trade strengths. Radar sees through darkness, sunglasses, and a hand over the face, and it can reach passengers a driver-facing camera cannot. Tesla, Vayyar, and Cipia all use radar for in-cabin sensing. A camera’s edge is economic: it is hardware the car already carries to satisfy monitoring regulation, so adding heart rate costs no new part. Smart Eye is betting that “already installed” outweighs radar’s technical advantages for automakers watching cost.
Could this stop a car if the driver has a heart attack?
Not today, and not as announced. Smart Eye described detection, not response, and used conditional language throughout: such a system “could eventually” help. A heart-rate reading only becomes a safety feature when the car is programmed to act on it, by warning, slowing, pulling over, or calling for help. Euro NCAP’s roadmap envisions in-cabin sensors reporting vital signs to emergency responders after a crash from 2026, which is where regulators want this to lead.
Is this a privacy problem?
It is a fair question. Physiological data is sensitive, and a camera continuously estimating your pulse raises legitimate concerns about what is stored and who sees it. European DMS rules already limit how driver-monitoring data may be retained and require that it stay within the vehicle’s safety functions. Any automaker adopting vital-signs monitoring would have to satisfy those data-protection rules, but buyers should expect to see the fine print before trusting it.
Does breathing detection work as well as heart rate?
Smart Eye’s own language suggests not. It says heart rate is “detected” while breathing rate is “estimated,” a wording gap that reflects a real difference in difficulty. Respiration has to be inferred from faint secondary effects on the pulse signal or from small chest and shoulder movements, both easily lost to the motion of driving. Heart rate is the more mature of the two metrics.
Who is Smart Eye?
A Swedish driver-monitoring and human-behavior AI company founded in Gothenburg in 1999, listed on Nasdaq First North. It acquired Affectiva and iMotions in 2021 and supplies software to automakers including Volvo, BMW, GM, Polestar, and Geely, with 357 automotive design wins across 22 brands and software in 75 car models as of late 2024. Its non-automotive customers include NASA and Boeing.
When and where is the live demonstration?
At InCabin USA, the in-cabin sensing conference held at Huntington Place in Detroit from June 9 to 11, 2026. It is the first public demonstration of Smart Eye’s camera-based vital-signs capability.
Reporting drawn from Smart Eye’s June 9, 2026 announcement and 2023 InCabin Brussels disclosure, the EU General Safety Regulation (2019/2144), Euro NCAP’s Vision 2030 roadmap, NHTSA crash-causation data, Tesla software-update documentation, and the SparsePPG in-vehicle rPPG study (2018). Accuracy figures cited are from independent research, not from Smart Eye, which published none.
Michael Kahn
Michael Kahn is the writer, photographer, and publisher behind The Weekly Driver. He cares about how cars drive and what they're like to own. He covers automobile industry news, car shows and events, and new car reviews. The reviews come from behind the wheel: day trips that favor back routes, treating a good meal as half the reason to go. He directs and produces the visual media, matching each car to a setting and mood that fit it. When he's not reviewing new cars, Michael races paddleboards, camels, and ostriches, along with the occasional exotic car on the racetrack, and has driven in every state and country visited.
https://theweeklydriver.com