Developing the next generation of eye tracking

Reliable / Accurate / Easy

The Challenge

Eye tracking is essential for augmented reality (AR) technology to realize its full potential and for the AR era to commence. True AR glasses need to know where users are looking – both direction and distance – accurately and reliably.

AR requires eye tracking that works with any user, under any condition, without the need for recalibration and with no sensitivity to slippage (device movements on users’ heads).

The Solution

Immersix developed the first commercial eye tracking using retinal imaging.

  • A single setup process lasts forever with no need for recalibration. As soon as a user puts on the glasses, the system tracks their gaze immediately and without bias.
  • Not sensitive to slippage.
  • Not sensitive to eye shape or color.
  • Not sensitive to ambient light conditions.
  • Better than 0.5 degrees accuracy and better than 0.1 degrees precision.
  • Automatic identification of users. A single device can serve multiple users with hassle-free user authentication.
  • Requires only a single small component on the frame front for each eye.

The Technology

At the heart of Immersix’s technology is a small sensor that combines a light source and a camera to capture IR images of the retina through the pupil.

  • Immersix developed algorithms that translate retinal images into eye orientations independently of eye position. This unique feature eliminates the slippage problem and enables bias-free gaze estimation.
  • The retina hardly changes over time so a single setup process lasts forever with no need for recalibration.
  • Every person’s retinal pattern is unique, enabling automatic identification of users.
  • Tracking retinal optical features is robust and accurate due to the rich details of the retina. Moreover, a person’s gaze is determined by the position of their fovea (a region on the retina that corresponds to sharp central vision). Thus, tracking retinal features is as close as you can get to tracking the gaze directly. The end result is a highly accurate and precise gaze estimation.
  • Retinal-imaging-based eye tracking needs no more than a view of the pupil, and is therefore not affected by eye shape or color.

Management Team

  • Ori
    Ori Weitz,
    Founder and CTO

    Algorithms and computer vision expert. Developed a novel positional tracking technology and a new method of position coding. Since late 2016 Ori has focused his attention on solving the gaze tracking problem.
    Ph.D. candidate, M.Sc. summa cum laude, B.Sc. magna cum laude.

  • Haim
    Haim Perski,

    More than 25 years of experience in the high-tech industry as well as specific expertise leading startup companies and R&D organizations in developing innovative products for new markets. He has served in a range of positions including CEO of FarmSee, CEO of PointGrab, and CTO of N-trig.
    Haim holds a B.Sc. in Electrical Engineering, Cum Laude, from Tel Aviv University.

  • Itay
    Itay Cohen,

    Over 28 years of technological experience in growing startups from inception to exit. Former director of hardware at Modu. Vast experience in building and managing multidisciplinary teams and projects (hardware, software, mechanical design, and industrial design) with a focus on consumer device development.

  • Saar
    Saar Wilf,

    Over 20 years as a serial entrepreneur and investor focusing on AI, Fintech and computer vision. Founded Fraud Sciences (acquired by PayPal), ClarityRay (acquired by Yahoo), Trivnet (acquired by Gemalto). First investor in Moon Active (valued at $5 billion), Supersonic (acquired by Iron Source), Crosswise (acquired by Oracle).