Imaging company Almalence has released a trial plugin for its Digital Lens technology which makes use of eye-tracking to purportedly increase the resolving power and clarity of XR headsets.

Almalense argues that the lenses on most XR headsets today aren’t being used to their fullest potential. By taking advantage of eye-tracking and smarter calibration, the company says its image pre-processing technology can actually increase the resolving power of a headset, including expanding the ‘sweet spot’ (the part of the lens with the highest visual fidelity).

The company has released a trial version of its technology through a plugin that works with Pico 3 Neo Pro Eye, HP Reverb G2 Omnicept, and HTC Vive Pro Eye. The plugin works with OpenXR compatible content, and even allows users to switch back and forth between each headset’s built-in image processing and the Almalence Digital Lens processing.

Based on through-the-lens demonstrations by the company, the technology does objectively increase the resolving power of the headsets. The company focuses on doing more advanced pre-processing to account for artifacts introduced by the lens, like chromatic aberration and image distortion. In essence the software increases the sharpness of the image by making the light passing through the lens land more precisely where it’s supposed to.

Almalence has shared heat maps comparing the changes in visual quality with and without its image technology, along with a broader explanation of how it works.

Another big advantage over the status quo, Almalence says, is the Digital Lens tech uses eye-tracking to perform these corrections in real-time, meaning that as you move your eyes around the scene (and off-axis from the center of the lens), the corrections are updated to account for the new angles. This can expand the ‘sweet spot’ of the lens and ‘pupil swim’ by making adjustments to account for the position of the pupil relative to the center of the lens. This video demonstrates the pupil swim correction:

The plugin, which anyone can use until January 2024, aims to demonstrate the company’s claims. Ultimately it appears the company wants to license its technology to headset makers to improve image quality out of the box.

Newsletter graphic

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. More information.

Ben is the world's most senior professional analyst solely dedicated to the XR industry, having founded Road to VR in 2011—a year before the Oculus Kickstarter sparked a resurgence that led to the modern XR landscape. He has authored more than 3,000 articles chronicling the evolution of the XR industry over more than a decade. With that unique perspective, Ben has been consistently recognized as one of the most influential voices in XR, giving keynotes and joining panel and podcast discussions at key industry events. He is a self-described "journalist and analyst, not evangelist."
  • Arno van Wingerde

    Well it sounds like they really went after a large market share: Pico 3 Neo Pro Eye, HP Reverb G2 Omnicept, AND HTC Vive Pro Eye! But eh… how about the Quest? I heart Meta sold quite a few of those….

    • jerronimo3000

      Requires eye tracking, so unless you mean quest pro, there’s not a lot of those out there either.

      Looks like their stuff may require a PC though given the supported headsets

      • Sven Viking

        Yeah, unfortunately Quest Pro doesn’t support various features including eye tracking for Link/AirLink, making it useless for certain business applications.

      • Eugene Panich

        Quest Pro’s eye-tracking does not provide the coordinates of the eye pupil (which the technology needs), only the gaze direction.
        Re PC – yes the solution is for tethered devices only, but we are in process of making it work on untethered Snapdragon-based devices as well.

        • Zig

          Are you going to enable this technology for Pimax Crystal? I think it could benefit a lot, crystal has big problems with CA

          • Eugene Panich

            Working on that right now :)

    • ViRGiN

      You don’t really think they can compete with Meta?

      And obviously they can’t just release it for Meta family

  • Csaba Szalontai

    Pimax Crystal?!?!?!?

  • That’s cool!

  • psuedonymous

    This is one of the big wins with eye-tracking (bigger than foveated rendering or UI interactions, IMO). I’ve always been surprised that Quest Pro and other eye-tracked HMDs did not implement this low hanging fruit of fidelity upgrade already.

    • Wonder if there are performance limitations keeping it from running on standalone chipsets yet. Unlike ETFR which trades off some performance gains in other areas, I think this would just stack onto the processing load.

      • psuedonymous

        It depends on how much of the compensation you can pre-compute. e.g. rather than computing a new pre-warp matrix on the fly for each frame, if you have (e.g.) a 10×10 grid of pre-computed matrices per eye, then each frame you only need to select the nearest and blend them to produce the new matrix to use that frame, trading off some fidelity for greater performance. This is also exactly the sort of thing that should eventually be offloaded to a FFB in your SoC to bring the performance impact down to effectively zero.

        Plus for clarity-increase-per-unit-performance it very likely beats out MSAA and SSAA!

        • Eugene Panich

          Good point. The “secret sauce” of the technology is that we managed to push most of compute-intensive processing to the calibration stage. The runtime part is very fast.

      • Eugene Panich

        There are performance challenges indeed, but we are positive that we will achieve short enough latency on untethered devices, at least those based on Qualcomm Snapdragon platforms.

  • Daniel Dobson

    I love the way that they use a robotic arm / eye for precision testing

  • asdpls

    I’m glad you reported on this. Things like pupil swim correction are super important. Beyond this, I wonder if they will address ocular parallax rendering and dynamic IPD, though you may need higher performance eye tracking for these things. Research is showing these two latter problems are as important for visual comfort as dynamic focus. See the research Stanford summarized in “Eye Tracking Revisited” on youtube (not going to even bother linking to anything on disqus)

  • JanO

    This is the first time I’m sad the Q3 doesn’t feature eye tracking as this tech seems promising…

    @Ben : Thanks for the great article and great journalism in general… A subject I’d like to read about is the ongoing graphical improvements regularly brought by Quest’s system updates… A few of them really made a big graphical quality jump, yet I’ve never seen anyone covering the subject exept for the recent cpu/gpu boost and corresponding resolution bump… I’m sure there are other factors at play here…

  • Eugene Panich

    Just released an update, improving stability and compatibility with various apps (including those with EAC).

  • Eugene Panich

    A version supporting Pimax Crystal has just been released, check it out :)