Add eye tracking superpowers to your product
NeonNet is an advanced eye-tracking technology designed for integration into OEM systems. It sets the industry standard for reliability and ease of use and is already employed across thousands of professional applications.
NeonNet can be integrated with off-the-shelf components or customized for specific hardware, making it suitable for both proof-of-concept projects and full product integrations.
Latency
<10 ms (photon to output)
Sampling rate
Up to 480 Hz
Sensor size
1.5 x 1.5 x 1.5 mm
Power consumption
<15 mW for the sensor system
Battery life
8+ hours
Platform compatibility
Qualcomm XR & Snapdragon, Nvidia GPU & Jetson
Real-world accuracy
1.8° uncalibrated; 1.3° with offset correction
Eye tracking FOV
100° x 80° covering the full human FOV
Easy to integrate
Tiny & low-power sensors
One tiny camera per eye, nothing else. As small as 1.5 x 1.5 x 1.5 mm and consuming 7.2 mW.
Sampling rate
Up to 480 Hz
Eye camera size
1.5 x 1.5 x 1.5 mm
Power consumption
<15 mW for the sensor system
Research-grade data
Comprehensive data
Signals include gaze (monocular & binocular), fixations, saccades, blinks, pupil diameter, eyeball pose, and eye openness.
Gaze accuracy
Great gaze accuracy at 1.8° uncalibrated and 1.3° with offset correction.
Complete FOV
NeonNet covers the full human field of view measuring gaze across 100° x 80°.
Substantially more reliable
Population coverage
No matter the gender, age, or ethnic background of the user, NeonNet ensures reliable eye tracking across the entire population.
Adverse lighting
NeonNet is invariant to lighting conditions, ensuring accurate tracking indoors, outdoors, and in low-light environments.
Movement and slippage
NeonNet remains accurate even when the user moves dynamically or displaces the eyewear.
The only eye tracking technology requiring no calibration or setup
No person-specific setup
No setup or tuning for a specific user. NeonNet works out of the box for anyone.
No session calibration
It works as soon as you put it on. No calibration is required ever.
Intuitive experience
With no instructions or learning curve, NeonNet eliminates user error and ensures a foolproof experience.
Proven performance in diverse applications
The Neon Eye Tracking Module is one concrete embodiment of NeonNet. It has become the de facto standard tool in numerous fields of academic research.
Elite sports teams, medical doctors, neuromarketers, professional performance trainers, and leading scientists are successfully using NeonNet daily.
Revolutionizing eye tracking with machine learning
NeonNet represents a cutting-edge approach to eye tracking, blending traditional camera-based techniques with advanced machine learning to overcome the longstanding challenges of reliability and usability in eye tracking.
Similar to traditional methods, NeonNet employs geometric modeling of the eyes. It utilizes an introspectable pipeline grounded in physiological principles, ensuring transparency and scientific validity.
By incorporating machine learning, NeonNet massively enhances the system's robustness, effectively making it invariant to things like slippage and lighting conditions.
This unique and patented combination of geometric modeling and machine learning eliminates the need for calibration while consistently delivering research-grade data quality.
By avoiding a dependency on volatile features like glints or particular features of the eye, and instead relying on simple camera technology, NeonNet is applicable in virtually every scenario.
Ready to integrate NeonNet?
Module integration
Get started immediately by integrating the existing Neon module and Neon Companion device into your setup. With open-source interface geometry and data formats, prototyping is very fast and cost-effective.
Timeline: Days
Our sensors, your hardware
Position our sensors correctly in your device, and receive a NeonNet library optimized for your SOC. This setup has minimal form factor requirements and allows for deep and streamlined integration.
Timeline: 1-2 months
Fully customized solution
Choose your own sensor configuration. Pupil Labs will customize NeonNet to your specifications and validate its performance to meet your unique requirements.
Timeline: 3-12 months
Why Pupil Labs?
Pupil Labs has been innovating eye tracking technology for more than 10 years. We began with a focus on applications in scientific research and have since expanded to meet the challenges of real-world applications.
Trusted by experts
Our users are spanning elite sports teams, professional trainers, medical doctors, neuromarketers, leading scientists, astronauts, and more.
10,000+ Users
Today, our technology is trusted by 10,000+ users worldwide, including most major universities and Fortune 500 companies.
Extensive patent portfolio
Pupil Labs products are backed by an extensive IP portfolio and driven by a diverse team of experts committed to continuous innovation.