Smart Eye is a Swedish company, founded in 1999 and headquartered in Gothenburg, that develops camera-based technology to understand human attention, state and behavior in real time. The technology is used in millions of vehicles, in research environments, and in products across industries — and, through Sightic, in impairment screening for police and safety-critical settings. 

It’s the plain-language way we describe what our technology does: interpret human visual and behavioral signals — attention, state and intent — in real time, and translate them into reliable data that systems can act on. 

Human Insight AI is Smart Eye’s proprietary technology for interpreting human visual and behavioral signals in real time. It combines head, eye and gaze tracking, facial expression analysis, activity and object detection, and multimodal data fusion to turn camera input into a reliable, structured understanding of human attention, state and behavior. 

Across automotive (driver monitoring and interior sensing), behavioral research, training and simulation, product integration, and impairment screening — anywhere understanding human behavior improves safety, performance or reliability. 

Eye tracking captures and measures a person’s gaze and eye movements to generate insight into their attention, alertness and focus. Sensors detect the eye, and AI interprets the data in real time — including in demanding conditions where lighting varies or the eyes are partly covered. 

Infrared light creates a reflection in the eye’s cornea that a camera can capture. Our systems track each eye separately, identify the iris and pupil, and combine the two feeds into a single consensus gaze — while also tracking facial features and head movement, so head pose stays accurate even when the face is partly obscured or lighting is poor. 

Reliability comes down to accuracy and gaze precision. Under ideal conditions, Smart Eye Pro delivers head and gaze accuracy better than 0.5°. Because our remote systems track each eye separately and use multiple cameras, performance holds up when lighting varies, the face is partly obscured, or there’s movement — in vehicles, simulators and control rooms. 

It depends on your environment and study. Smart Eye Pro suits demanding, large-scale setups such as simulators and multi-screen research; XO adds dual-camera durability; and Aurora is compact and easy to move between projects and to synchronize with sensors like EEG. Our team can help match a system to your work. 

We treat ethical AI as a priority. Our training data is deliberately diverse across gender, ethnicity and age, and where personal data is involved we design for privacy from the start — minimizing storage and, where possible, processing data on-device. 

Yes. Our software runs on standard hardware and integrates into existing products, apps and workflows — from a vehicle platform to a rental app’s unlock flow or a police system — as a standalone app or an SDK, with custom integrations scoped to your setup. 

Pricing depends on what you’re solving — the products, configuration and scale involved — so quotes are tailored. Tell us what you’re working on and our team will point you to the right solution and next steps. 

A DMS uses an in-car camera, computer vision and AI to read a driver’s eyes, face and head movement and measure alertness, attention and focus in real time. By detecting distraction, drowsiness and impairment, it helps prevent accidents and improve road safety. 

Distraction, drowsiness and attention, plus features such as driver identification, dangerous-behavior detection, object detection, seatbelt detection, and — through Affectiva’s facial coding — facial expression and state. It works reliably across lighting conditions, and for faces with masks or glasses. 

 Smart Eye’s driver monitoring software is in more than four million vehicles on the road, across models from manufacturers worldwide — and the number continues to grow. 

Interior Sensing extends monitoring from the driver to the whole cabin, analyzing the state, behavior and position of every occupant — and the objects they use. It supports safety features and enables comfort and wellbeing features that adapt to the people in the car. 

Occupancy and occupant state, face recognition, child or pet left behind, child-seat and object detection, activities such as phone use, and — through facial coding — facial expression and mood, for every occupant. 

It helps prevent tragedies like a child or pet left in a hot car, supports passive-safety functions such as adaptive airbag deployment and seatbelt use, and — with a complete view of the cabin — helps understand what’s pulling a driver’s attention from the road. 

Yes. Our fleet and aftermarket system uses the same automotive-grade driver monitoring software proven in millions of cars, in a system that’s straightforward to install in trucks, buses and passenger cars. It detects drowsiness and distraction to improve safety on every trip, and integrates with existing vehicle systems. 

Yes. Our driver monitoring software meets Euro NCAP, the EU General Safety Regulation (GSR) and China’s GB/T standards for driver monitoring. As legislation expands from monitoring attention toward addressing impaired driving, we’re extending the same core technology into camera-based detection of alcohol and drug impairment. 

A fast, camera-based way to check whether someone shows signs of impairment — reading eye movements and ocular responses that drugs and alcohol reliably alter, and returning an objective result in seconds on a standard smartphone. It’s used by police at the roadside and by employers before a shift. This technology comes from Sightic, now part of Smart Eye. 

Drugs produce involuntary changes in eye movements and ocular responses that can’t be masked. A short scan captures those signals and analyzes them on-device — no swabs, no biological samples. Only positive screens proceed to further verification. 

Drug screening takes about 15 seconds and alcohol screening around 6 seconds, both on a standard smartphone — no dedicated hardware. 

It’s a screening and decision-support tool. It gives an objective, consistent basis for deciding whether a case needs further testing, such as a blood analysis — it doesn’t replace evidential testing. 

At a 1% false-positive rate, the alcohol screening correctly identifies 97.2% of clearly impaired people at 0.8‰ BAC and above — the level where crash risk is roughly ten times higher than for a sober person. 

Iris recognition confirms who is actually present in under a second, so only authorized people enter secure areas or operate vehicles and machinery. It’s the same iris capability Smart Eye uses elsewhere, applied to access. 

Yes. All processing happens on-device, no images are stored and no biometric data is retained. The solutions are GDPR-compliant and ISO 27001 certified, with formal ethics approval behind the models. 

Eye tracking gives researchers objective data on how people perceive, think and respond — across automotive, aviation, psychology and neuroscience, media and marketing, and many other fields. Hundreds of universities and companies use our technology, including NASA, Airbus, Boeing, Volvo and Harvard University. 

It combines several data sources — eye tracking, facial expression, EEG, ECG, EMG, galvanic skin response and heart rate — into one synchronized view. Because much of behavior happens below conscious awareness, combining signals gives a fuller, more reliable picture than any single measure. 

Yes. Through the iMotions platform, eye tracking synchronizes with facial coding, EEG, ECG, EMG, galvanic skin response and heart rate, so you can capture emotional, cognitive and behavioral data together in one environment. 

By tracking gaze, head movement, pupillometry and eyelid behavior, our systems support human-factors research, pilot monitoring and pilot training — helping instructors teach and correct scanning technique, on the ground and in the air. NASA uses our technology in its flight deck simulators. 

Inadequate cockpit scanning is a factor in nearly half of commercial aviation incidents. Used in simulators, eye tracking makes scanning behavior visible — so problems can be caught and corrected before they become incidents. 

It shows instructors where and when a trainee looks — gaze, scan path and attention over time — giving objective evidence of how information is perceived, prioritized and acted on. It’s used in aviation, air traffic control, automotive, rail and control-room training, and supports Evidence-Based Training (EBT) and Competency-Based Training and Assessment (CBTA). 

No. Tracking is fully remote — no headgear or attachments — so trainees behave naturally and instructors see performance as it really occurs. 

Using Affectiva’s facial coding, we measure consenting viewers’ unfiltered emotional and cognitive responses to ads, film and content — at scale. It helps brands and researchers understand engagement and answer creative, media and ROI questions, even when people can’t or won’t put their response into words. 

Yes. It captures authentic emotional responses from participants — in person or online — adding a layer of insight beyond what people self-report, and complementing interviews and concept development. 

iMotions is Smart Eye’s multimodal platform for human behavior research, founded in 2005 and based in Copenhagen. It combines data from many sensors — eye tracking, facial expression, EEG, GSR and more — into one environment that captures, visualizes and analyzes emotional, cognitive and behavioral data in real time. It’s part of the Smart Eye Group and runs as its own company. 

iMotions is hardware-agnostic. It integrates eye trackers along with EEG, ECG, EMG, GSR/EDA and facial expression analysis, and partners with leading sensor manufacturers for straightforward setup. 

Affectiva, now part of Smart Eye, pioneered facial coding — reading emotion and cognitive state from facial expressions, the field it named Emotion AI. It spun out of the MIT Media Lab in 2009, founded by Dr. Rana el Kaliouby and Dr. Rosalind Picard, and joined Smart Eye in 2021. Its technology detects nuanced emotions and cognitive states and powers Smart Eye’s interior sensing and media analytics work. 

Using a standard camera, Affectiva’s AI detects a face in real time, identifies key facial features, and classifies expressions with deep learning — mapping them to emotions and to cognitive states such as fatigue, attention, confusion and distraction. The models are trained on naturalistic face videos from more than 90 countries and designed to perform fairly across demographic groups. 

In vehicles, it adds emotional and cognitive context to interior sensing — reading the state of the driver and occupants. In research, it measures how people respond to content, unobtrusively and at scale. 

Yes. Facial expression analysis is based on validated academic research and, conducted with consent, is an accurate and ethical way to understand human responses. We treat context as essential to avoid simplistic conclusions, and encourage combining signals from multiple sources for the fullest picture. 

Sightic, now part of Smart Eye, develops camera-based impairment detection — screening for drug and alcohol impairment from eye and facial signals on a standard smartphone. It powers Smart Eye’s impairment screening for police and safety-critical settings. 

Yes. Smart Eye has been publicly listed since 2016 and trades in Stockholm. Reports, presentations, the financial calendar and governance information are available on our investor pages. 

Financial reports, presentations, the financial calendar, share and governance information, and investor-relations contact details are all on our investor pages.