Toggle menu
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

Dynamic FOV Reduction

vrarwiki.com · VR & AR Wiki
(Redirected from FOV Reduction)

Dynamic FOV Reduction, also sometimes referred to as Blinders, Blinkers, or Vignetting, is a user interface feature some VR creators offer in their VR content. In an effort to reduce potential motion sickness, when the user performs certain in-game motion actions, such as turning, running, jumping, falling, driving, etc, their FOV is automatically temporarily reduced. This may help reduce or prevent motion sickness.[1]

Developers and researchers also call the technique "tunneling", "tunnel vision" or dynamic "field-of-view restriction".[2][3][4] It targets VR sickness caused by artificial locomotion, where the eyes see the body moving through the virtual world while the vestibular system senses that it is still. Because the visual sense of self-motion (vection) is driven most strongly by optical flow in the periphery of the retina, hiding the periphery while the user moves reduces the visual-vestibular conflict.[5][6] The cost is a smaller visible area, which can reduce the sense of presence and hide parts of the environment.[1][7]

Reviewed 27 September 2026. Checked the Fernandes and Feiner 2016 paper, the USC patent dates, the eight research abstracts in the table, the Meta, Unity and Google developer documentation, and the Eagle Flight and Google Earth VR press sources against the new text. About review dates.

How it works

The user's view might be restricted to a small circle, or have black borders on the left and right, or the screen may be briefly blacked-out entirely.

In some VR content, this is a feature that can be turned on or off, or adjusted as to when or how often it will occur, and how significantly the FOV will be altered. Setting a FOV Reduction setting to a "strong" setting would mean that the user's FOV is significantly reduced when the reduction is triggered.

The standard implementation is a symmetric mask that covers the periphery in response to the user's translational speed, angular velocity, or both.[8] The mask usually has a soft edge. In the 2016 study by Ajoy S. Fernandes and Steven K. Feiner, pilot participants found hard-edged cutouts distracting and saw Mach bands when the gradient between the clear and opaque zones was narrow, so the researchers chose a cutout whose transparency fell off linearly across a wide band.[1] Google's tunneling demo for developers had three parts: the environment with a restricted field of view, the "iris" (the opening through which the environment is seen), and a static "cage" background in the user's peripheral vision that gives a fixed frame of reference.[9] Rob Jagnow of Google's Daydream Labs compared the effect to watching television, where the screen fills only a small part of the visual field and a stationary room grounds the periphery, and noted that fading the tunnel in and out keeps it from being distracting.[2]

Implementations differ in what triggers the reduction. Meta's Interaction SDK has a LocomotionTunneling component that can narrow the view when the user rotates above a set speed, moves above a set linear speed, or accelerates and decelerates; its vignette is defined in degrees so that the amount of occlusion stays the same whatever the headset's real field of view.[10] Meta's design guidance says vignettes "darken or completely occlude the edges of the screen when movement occurs", warns that a smaller field of view can feel disorienting or claustrophobic, and describes more advanced versions that analyze scene geometry and hide only the areas carrying the most optic flow.[3]

Parameters in common toolkits

Toolkit Component Adjustable settings
Unity XR Interaction Toolkit Tunneling Vignette Controller Aperture size, feathering effect, ease-in time, ease-out time, ease-in time lock, ease-out delay time, vignette color, vignette color blend, aperture vertical position[11]
Meta Interaction SDK LocomotionTunneling (using TunnelingEffect) Fade-out speed, feathering, color, allowed field of view for rotation, movement and acceleration events[10]
Google VR (Daydream Elements) Tunneling demo Fade speed, iris speed, maximum speed, maximum angular velocity, smoothing factor, cage color, preview field of view (degrees), preview alpha[9]

Unity added a Tunneling Vignette sample to the XR Interaction Toolkit in version 2.1, describing the vignette as "a comfort mode intended to mitigate motion sickness in VR"; the first preview of that version, 2.1.0-pre.1, is dated 2 May 2022.[12] The component renders a hemisphere with a transparent inner aperture and a black cut-off, and is placed as a child of the main camera. Locomotion providers trigger it according to their locomotion phase.[11]

History

Researchers had long studied the relationship between display field of view, presence and simulator sickness, but most of that work used a single field of view for a whole session. Kim and colleagues built a system for a multi-projector environment that monitored biosignals and narrowed the field of view in discrete steps when they indicated cybersickness.[1]

Mark Bolas, J. Adam Jones, Ian McDowall and Evan Suma filed a US patent application titled "Dynamic field of view throttling as a means of improving user experience in head mounted virtual environments" on 17 March 2014, claiming priority to 15 March 2013. It was published on 18 September 2014 as US 2014/0268356 A1 and granted on 9 May 2017 as US 9,645,395; the current assignee is the University of Southern California. The patent describes an HMD connected to a sensor that monitors the user's physiological state so that simulator sickness can be detected or predicted and the field of view throttled.[13] Fernandes and Feiner described the application as the closest work to their own, while noting that it did not report a user study.[1]

At the IEEE Symposium on 3D User Interfaces in March 2016, Fernandes and Feiner of Columbia University presented what they described as the first study of automatically and continuously changing a participant's field of view throughout a VR session.[1] Their testbed used an Oculus Rift DK2 and a modified version of the Oculus Tuscany demo built in Unity 4. Seated participants moved with a gamepad, and the restrictors contracted as a function of gamepad translation speed and angular velocity; physical head motion did not trigger them. Of the 30 participants whose data were analyzed, 15 used a minimum field of view of 90 degrees and 15 a minimum of 80 degrees. The authors reported that the restrictors helped participants stay in the environment longer and feel more comfortable than in the control condition, without a significant drop in presence scores. Half of the participants (15 of 30) gave the lowest rating ("Did not notice or did not happen") when asked whether they had noticed the field of view shrinking, and 11 of the 15 participants in the 90-degree group did so.[1]

Commercial titles adopted the technique around the same time. Ubisoft's flying game Eagle Flight used what UploadVR called "dynamic blinders".[14] In a talk at the VRDC conference in San Francisco in 2016, Olivier Palmieri of the Eagle Flight team explained that the edges of the player's view darken during high-risk maneuvers such as fast turns and fast flight close to large objects, hiding the peripheral vision that is most sensitive to motion.[15] Palmieri said that people were skeptical of the blinders when they were first shown, but that players testing the prototype "didn't even notice."[14] Google Earth VR launched for the HTC Vive on 16 November 2016 with a comfort mode that narrowed the field of view while the user flew; the mode could be switched off.[16][17] Meta's current locomotion design guidance cites Eagle Flight as an example of advanced vignetting.[3]

By the early 2020s researchers described dynamic field-of-view restriction as a technique "widely implemented in commercial VR games".[7] Game engines and platform SDKs now ship ready-made versions, as the table above shows.

Research

Study Venue and year Participants Main finding
Fernandes and Feiner[1] IEEE 3DUI, 2016 30 Speed-driven soft-edged restrictor reduced discomfort in first sessions; no significant loss of presence; most participants in the 90-degree condition did not notice it
Al Zayer, Adhanom, MacNeilage and Folmer[6] ACM CHI, 2019 28 (14 women) Restrictor mitigated VR sickness in both sexes; no negative effect on spatial navigation performance found
Adhanom, Griffin, MacNeilage and Folmer[5] IEEE VR, 2020 22 Gaze-following (foveated) restrictor versus head-fixed restrictor: no significant difference in sickness or noticeability; wider eye-gaze scan area with the foveated version
Teixeira and Palmisano[4] Virtual Reality (journal), 2020 online, 2021 issue 40 In Marvel Powers United VR, dynamic FOV restriction significantly reduced cybersickness
Adhanom, Al Zayer, MacNeilage and Folmer[18] ACM Transactions on Applied Perception, 2021 28 (14 men, 14 women) Restrictor equally effective in both sexes; did not impede spatial learning
Shi, Liang, Wu, Yu and Xu[19] Proc. ACM Comput. Graph. Interact. Tech., 2021 32 In a racing game, no significant differences in sickness, presence or workload between FOV reduction, depth-of-field blur and a rest frame; FOV reduction caused information loss
Wu, Bailey, Stoffregen and Suma Rosenberg[7] ACM SUI, 2021 Remote sample of VR headset owners Asymmetric restrictor that keeps the ground visible improved comfort, postural stability and presence
Wu and Suma Rosenberg[8] ACM VRST, 2022 38 Adaptive restrictor driven by real-time optical flow reduced cybersickness versus control; higher presence than symmetric restriction

Results have not been uniform. Teixeira and Palmisano had 40 participants play Marvel Powers United VR for 10-minute sessions with and without dynamic restriction over three sessions, and found that cybersickness rose steadily during each exposure and was significantly reduced by the restriction; they concluded that it "can serve as a viable countermeasure to cybersickness".[4] Shi and colleagues, using the Fernandes and Feiner method in a racing game, found no significant benefit over other techniques. Some participants in their high-control condition reported discomfort and disorientation from frequent changes in the size of their view, particularly after collisions, and the authors suggested that dynamic reduction may not suit scenarios with frequent changes in linear or angular acceleration.[19]

Other studies have changed the shape or placement of the restrictor. Nahal Norouzi, Gerd Bruder and Greg Welch examined vignetting as a way to reduce sickness during amplified head rotations, a case where the view turns faster than the head.[20] Adhanom and colleagues moved the center of the restrictor with the user's gaze using eye tracking, because with a head-fixed restrictor a user who looks off-center can look at the mask itself and be exposed to peripheral optical flow.[5] Fei Wu and Evan Suma Rosenberg proposed restrictors that keep the ground plane visible, and later an adaptive restrictor with an asymmetric mask that hides only the peripheral regions with higher optical flow.[7][8]

See also

References

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Ajoy S. Fernandes, Steven K. Feiner (2016-03). "Combating VR Sickness through Subtle Dynamic Field-Of-View Modification". 2016 IEEE Symposium on 3D User Interfaces (3DUI), pp. 201-210. Columbia University. doi:10.1109/3DUI.2016.7460053. http://www.cs.columbia.edu/2016/combating-vr-sickness/images/combating-vr-sickness.pdf. Retrieved 2026-09-27.
  2. ↑ 2.0 2.1 Rob Jagnow (2017-06-06). "Daydream Labs: Locomotion in VR". The Keyword. Google. https://blog.google/products/google-ar-vr/daydream-labs-locomotion-vr/. Retrieved 2026-09-27.
  3. ↑ 3.0 3.1 3.2 "Reduce Optic Flow". Meta Horizon OS Developers. Meta. 2026-05-11. https://developers.meta.com/horizon/resources/locomotion-design-reduce-optic-flow/. Retrieved 2026-09-27.
  4. ↑ 4.0 4.1 4.2 Joel Teixeira, Stephen Palmisano (2020-08-12). "Effects of dynamic field-of-view restriction on cybersickness and presence in HMD-based virtual reality". Virtual Reality, vol. 25, no. 2, pp. 433-445. Springer. https://doi.org/10.1007/s10055-020-00466-2. Retrieved 2026-09-27.
  5. ↑ 5.0 5.1 5.2 Isayas Berhe Adhanom, Nathan Navarro Griffin, Paul MacNeilage, Eelke Folmer (2020-03). "The Effect of a Foveated Field-of-view Restrictor on VR Sickness". 2020 IEEE Conference on Virtual Reality and 3D User Interfaces (VR), pp. 645-652. IEEE. https://doi.org/10.1109/VR46266.2020.00087. Retrieved 2026-09-27.
  6. ↑ 6.0 6.1 Majed Al Zayer, Isayas Berhe Adhanom, Paul R. MacNeilage, Eelke Folmer (2019). "The Effect of Field-of-View Restriction on Sex Bias in VR Sickness and Spatial Navigation Performance". Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems. ACM. https://doi.org/10.1145/3290605.3300584. Retrieved 2026-09-27.
  7. ↑ 7.0 7.1 7.2 7.3 Fei Wu, George Bailey, Thomas A. Stoffregen, Evan Suma Rosenberg (2021-10). "Don't Block the Ground: Reducing Discomfort in Virtual Reality with an Asymmetric Field-of-View Restrictor". Symposium on Spatial User Interaction (SUI '21). ACM. https://doi.org/10.1145/3485279.3485284. Retrieved 2026-09-27.
  8. ↑ 8.0 8.1 8.2 Fei Wu, Evan Suma Rosenberg (2022-11). "Adaptive Field-of-view Restriction: Limiting Optical Flow to Mitigate Cybersickness in Virtual Reality". 28th ACM Symposium on Virtual Reality Software and Technology (VRST '22). ACM. https://doi.org/10.1145/3562939.3565611. Retrieved 2026-09-27.
  9. ↑ 9.0 9.1 "Tunneling demo". Google VR. Google for Developers. https://developers.google.com/vr/elements/tunneling. Retrieved 2026-09-27.
  10. ↑ 10.0 10.1 "Locomotion Comfort". Meta Horizon OS Developers. Meta. 2025-11-06. https://developers.meta.com/horizon/documentation/unity/unity-isdk-locomotion-comfort/. Retrieved 2026-09-27.
  11. ↑ 11.0 11.1 "Tunneling Vignette Controller". XR Interaction Toolkit 3.0 manual. Unity Technologies. https://docs.unity3d.com/Packages/[email protected]/manual/tunneling-vignette-controller.html. Retrieved 2026-09-27.
  12. ↑ "Changelog - XR Interaction Toolkit 2.1". Unity Documentation. Unity Technologies. https://docs.unity3d.com/Packages/[email protected]/changelog/CHANGELOG.html. Retrieved 2026-09-27.
  13. ↑ "US9645395B2 - Dynamic field of view throttling as a means of improving user experience in head mounted virtual environments". Google Patents. Google. https://patents.google.com/patent/US9645395B2/en. Retrieved 2026-09-27.
  14. ↑ 14.0 14.1 Ashley Whitlatch (2016-12-10). "Tunnel Vision: How Ubisoft Created 'Eagle Flight', A VR Flying Game With No Nausea". UploadVR. https://www.uploadvr.com/how-ubisoft-created-eagle-flight-sickness/. Retrieved 2026-09-27.
  15. ↑ Alex Wawro (2016-11-03). "Eagle Flight dev shares lessons learned about making comfy VR games". Game Developer. https://www.gamedeveloper.com/design/-i-eagle-flight-i-dev-shares-lessons-learned-about-making-comfy-vr-games. Retrieved 2026-09-27.
  16. ↑ Joe Durbin (2016-11-16). "'Google Earth VR' Launches For Free To Let You Fly Around The World". UploadVR. https://www.uploadvr.com/google-earth-vr-free/. Retrieved 2026-09-27.
  17. ↑ Frank Taylor (2016-11-16). "Google Releases Google Earth VR for HTC Vive". Google Earth Blog. https://www.gearthblog.com/blog/archives/2016/11/google-releases-google-earth-vr-htc-vive.html. Retrieved 2026-09-27.
  18. ↑ Isayas Berhe Adhanom, Majed Al Zayer, Paul R. MacNeilage, Eelke Folmer (2021-04). "Field-of-View Restriction to Reduce VR Sickness Does Not Impede Spatial Learning in Women". ACM Transactions on Applied Perception. ACM. https://doi.org/10.1145/3448304. Retrieved 2026-09-27.
  19. ↑ 19.0 19.1 Rongkai Shi, Hai-Ning Liang, Yu Wu, Difeng Yu, Wenge Xu (2021-05). "Virtual Reality Sickness Mitigation Methods: A Comparative Study in a Racing Game". Proceedings of the ACM on Computer Graphics and Interactive Techniques, vol. 4, no. 1. ACM. https://doi.org/10.1145/3451255. Retrieved 2026-09-27.
  20. ↑ Nahal Norouzi, Gerd Bruder, Greg Welch (2018-08). "Assessing Vignetting as a Means to Reduce VR Sickness during Amplified Head Rotations". Proceedings of the 15th ACM Symposium on Applied Perception (SAP '18). ACM. https://doi.org/10.1145/3225153.3225162. Retrieved 2026-09-27.