Toggle search
Toggle menu
5.4K
6.1K
230
52.9K
VR & AR Wiki
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

Light field

vrarwiki.com · VR & AR Wiki
More actions
(Redirected from Light Field)
See also: Terms and Technical Terms

A light field (also spelled lightfield) is a fundamental concept in optics and computer graphics that describes the amount of light traveling in every direction through every point in space.[1] In geometrical optics the quantity carried along each ray is radiance, measured in watts per steradian per square meter, so a light field can be treated as a function that gives the radiance of every ray according to its position and direction.[1] The concept was defined by Andrey Gershun in a 1936 paper, and the four-dimensional (4D) light field used in imaging and graphics was introduced to computer graphics in 1996 by Marc Levoy and Pat Hanrahan and, under the name Lumigraph, by Steven Gortler and colleagues.[2][3]

Light fields matter to virtual reality (VR) and augmented reality (AR) in three ways. A captured light field can be resampled to produce new, perspective-correct views of a real scene without building a 3D model, so a viewer can move the head and see parallax, reflections and other view-dependent effects; Google released a free VR demo of this, Welcome to Light Fields, in 2018.[1][4] A light field recorded through a microlens array can also be refocused after capture.[5] Finally, a display that reproduces the directions of light rays, rather than one flat image per eye, can give the eye correct focus cues, which is one of the main proposed remedies for the vergence-accommodation conflict in head-mounted displays.[6][7]

Reviewed 13 October 2026. Crossref metadata for every cited DOI, claims checked against Levoy 2006 and Adelson-Bergen 1991 full texts, project pages and press sources, the restored plenoptic wording and CREAL bullet, and October 2026 status of CREAL, JPEG Pleno, MIV and Google Beam. About review dates.

History

Early theory

Michael Faraday first speculated in 1846 in his lecture "Thoughts on Ray Vibrations" that light should be understood as a field, similar to the magnetic field he had studied; according to Levoy, Faraday was an experimentalist and could not formalize the idea, which James Clerk Maxwell's equations later did.[8][1] The term "light field" (svetovoe pole in Russian) was more formally defined by Andrey Gershun in a classic 1936 paper on the radiometric properties of light in three-dimensional space, written with the study of surface illumination by artificial lighting in mind.[1] An English translation appeared in 1939 in the Journal of Mathematics and Physics.[9] Gershun treated the light passing through a point as a sum of vectors, one per incoming direction with length proportional to radiance; integrating them gives the total irradiance at the point and a resulting direction, a quantity later called the vector irradiance field in computer graphics. Working before digital computers, he could not measure a light field, but he derived closed-form illumination patterns on surfaces lit by sources of various shapes.[1] The physicist Parry Moon later called the four-dimensional form of the function the "photic field".[1]

Optical capture of directional light information is older than the term. In 1908 Gabriel Lippmann proposed integral photography, in which an array of small lenses placed in front of a photographic plate records a perspective view of the scene from each lenslet position.[10][1] Lens-array light field cameras and integral imaging displays both descend from this idea.[1][11]

The plenoptic function and image-based rendering

In the context of computer vision and graphics, the concept was further developed with the introduction of the 7D plenoptic function by Edward Adelson and James Bergen in 1991. They proposed it as a single function describing everything that can be seen, and used it to derive the basic measurements of early vision such as motion, color and orientation.[12] This function describes all possible light rays, parameterized by 3D position (x, y, z), 2D direction (θ, φ), wavelength (λ), and time (t).[12] In 1992 Adelson and John Wang described single-lens stereo with a "plenoptic camera", an early design of the kind later developed into hand-held light field cameras, in which a lens array is inserted between the sensor and the main lens.[13][1] Leonard McMillan and Gary Bishop presented "plenoptic modeling", an image-based rendering system, at SIGGRAPH 1995.[14]

Practical computational approaches often reduce the dimensionality. Two seminal papers in 1996, "Light Field Rendering" by Levoy and Hanrahan at Stanford University[2] and "The Lumigraph" by Gortler et al. at Microsoft Research[3], independently proposed using a 4D subset of the plenoptic function for capturing and rendering complex scenes without detailed 3D models.[15] Levoy and Hanrahan interpreted the input images as 2D slices of a 4D function, constructed new views in real time by extracting slices in other directions, and reported compressing light fields by more than 100:1 with little loss of fidelity.[16] They introduced the influential two-plane parameterization (2PP), which defines each ray by its intersection points with two planes.[16] This 4D light field representation forms the basis for most later light field capture and display work. Analyses of how densely a light field must be sampled followed, including the "plenoptic sampling" study of Chai and colleagues in 2000.[17][1]

Computational photography and commercial cameras

In the 2000s research moved toward capturing light fields photographically. Stanford built a Multi-Camera Array of 100 custom video cameras, described at SIGGRAPH 2005, which could act as a single camera with a very large synthetic aperture, as a high-speed camera, or as a high-resolution, high-dynamic-range mosaic camera.[18] In April 2005 Ren Ng and colleagues described a hand-held plenoptic camera that inserts a microlens array between the main lens and the sensor and computes photographs focused at different depths from a single exposure.[5] In a companion paper Ng showed that, in the Fourier domain, a photograph is a 2D slice of the 4D light field, and that the sharpness of refocused photographs increases linearly with directional resolution.[19] Ng's Stanford dissertation won the 2006 ACM Doctoral Dissertation Award, and he went on to found Lytro.[15]

Lytro announced its first consumer light field camera on 19 October 2011, priced at US$399 for an 8 GB model, with shipping planned for early 2012; the company described the sensor as capturing "11 million light rays".[20] The company later turned to VR and cinema with Lytro Immerge and Lytro Cinema, then announced in March 2018 that it was winding down; some of its staff joined Google.[21] The German company Raytrix, based in Kiel, makes light field cameras that place a lens array in front of the image sensor, for industrial inspection, fluid-flow measurement, plant analysis and microscopy.[22]

Selected milestones

Year Work Significance
1846 Faraday, "Thoughts on Ray-Vibrations" Proposes that light be interpreted as a field[8][1]
1908 Lippmann, integral photography Lens arrays record many perspective views on one plate[10]
1936 (English 1939) Gershun, "The Light Field" Coins the term and treats light radiometrically in 3D space[9][1]
1991 Adelson and Bergen Define the 7D plenoptic function[12]
1992 Adelson and Wang Single-lens "plenoptic camera" for depth[13]
1996 Levoy and Hanrahan; Gortler et al. 4D light field rendering and the Lumigraph[2][3]
2005 Stanford camera array; Ng et al. hand-held plenoptic camera Photographic capture of light fields, refocusing after capture[18][5]
2011 Lytro camera announced Consumer light field camera with refocusing after capture[20]
2013 Lanman and Luebke, near-eye light field displays Microlens-based head-mounted display with focus cues[6]
2015 Light field stereoscope; Lytro Immerge announced Factored near-eye light field display; light field VR video camera[23][24]
2018 Google Welcome to Light Fields Free VR app of panoramic light field stills[4]
2020 NeRF; Google DeepView light field video Neural scene representation; six-degrees-of-freedom light field video[25][26]
2021 JPEG Pleno light field coding (ISO/IEC 21794-2) International standard for light field image coding[27]
2023 Meta Flamera light field passthrough prototype Perspective-correct passthrough without reprojection[28]

Theory and representation

The core idea behind the light field is to capture not just the intensity of light arriving at a point (like a conventional camera), but also the direction from which that light is arriving.[5]

The plenoptic function

The most complete representation is the 7D plenoptic function, P(x, y, z, θ, φ, λ, t), describing the radiance of light at any 3D point (x, y, z), in any direction (θ, φ), for any wavelength (λ), at any time (t).[12] Adelson and Bergen wrote the viewing position as (Vx, Vy, Vz) and named the function from the Latin plenus ("complete" or "full") and "optic". They described it as equivalent to a complete holographic movie of the visual world that would contain every photograph that could be taken of a region of space and time, neglecting the polarization and instantaneous phase of the light.[12] Because rotating an eye without moving it does not change the light reaching the pupil, the function needs no parameters for gaze direction.[12] For many applications, the full function is overly complex and contains redundant information: in free space, radiance does not change along a straight ray unless something blocks it.[1]

Simplified light fields

For static scenes under constant illumination, the time (t) dependency can be dropped, and wavelength is usually reduced to three color channels (red, green, blue). Levoy describes the radiance along all rays in a region illuminated by an unchanging arrangement of lights as a 5D function of position (x, y, z) and direction (θ, φ).[1] Outside the convex hull of the objects in a scene, radiance along a ray is constant from point to point, so exactly one of those five dimensions is redundant.[1]

  • 5D light field: L = L(x, y, z, θ, φ), the plenoptic function of a static scene.[1]
  • 4D light field: the most common simplification, formally defined as radiance along rays in empty space. Parry Moon called it the photic field, Levoy and Hanrahan the 4D light field, and Gortler et al. the Lumigraph.[1][3] It captures radiance along rays without the redundant dimension.[2]

A 4D light field can be understood as a 2D collection of 2D images of a scene, that is, a 4D array of pixels. Computing a new perspective view amounts to extracting an appropriately positioned and oriented 2D slice from this array.[1]

Two-plane parameterization (2PP)

This popular 4D parameterization defines a light ray by its intersection points with two planes, usually denoted the (u, v) plane and the (s, t) plane. A ray is thus identified by the coordinates L(u, v, s, t).[16] Levoy and Hanrahan used (u, v) for the camera plane and (s, t) for the focal plane.[16] A two-plane light field can be thought of as a collection of perspective images of the (s, t) plane, each taken from an observer position on the (u, v) plane, which relates it closely to the geometry of perspective imaging.[1] This representation is convenient because:

  • It relates well to how light field cameras with microlens arrays capture data: the lenslet positions sample one plane and the pixels behind each lenslet sample directions.[1]
  • It simplifies rendering algorithms, which often involve resampling this 4D function.[16]

The parameterization cannot represent every ray; for example, when the two planes are parallel, rays parallel to them are missing.[1]

Other 4D parameterizations exist, such as points on a plane or curved surface combined with the directions leaving each point, or pairs of points on the surface of a sphere.[1]

Sampling and resolution trade-offs

A light field is always sampled rather than recorded completely. If too few images are captured, renderings show ghosting from blended views; if the images have too few pixels, renderings are blurry, especially far from the original viewpoints. Adding an approximate 3D model of the scene reduces the number of images needed.[1] In a microlens camera the same sensor pixels must record both position and direction. Levoy gives the example of a prototype with a 16-megapixel sensor, an f/4 main lens and 300 x 300 microlenses: it could compute views of only 300 x 300 pixels, but could refocus anywhere within the depth of field of an f/56 camera.[1] This trade-off between angular and spatial resolution is a recurring constraint for both cameras and displays.[29][7]

Light field capture

Main article: Light field capture

Capturing a light field involves sampling the intensity and direction of light rays within a scene. Several methods exist:

  • Light field cameras (plenoptic cameras): These devices typically insert a microlens array between the main lens and the image sensor.[13][5] Each microlens samples the light arriving from the main lens from slightly different perspectives, directing these samples onto different pixels on the sensor below it. The sensor thus records not only spatial information but also angular information.[1] Lytro sold consumer models before shutting down in 2018, and Raytrix sells industrial models.[21][22]
  • Camera arrays: A synchronized array of conventional cameras can capture a light field by recording the scene from slightly different viewpoints simultaneously. Stanford's array of 100 video cameras is an early example.[18] Google's DeepView system recorded light field video for VR with 46 time-synchronized cameras on a 92 cm hemispherical dome.[26]
  • Scanning and gantry setups: A single camera moved precisely to multiple positions can sequentially capture the views needed to sample the light field. This is suitable for static scenes but slower. Levoy and Hanrahan digitized light fields with a video camera on a computer-controlled gantry, and Google's rig for Welcome to Light Fields swung a vertical arc of 16 cameras around a rotating platform for about a minute to record roughly 1,000 viewpoints.[16][4]
  • Coded aperture and mask-based photography: Placing an attenuating mask in the optical path of a camera can modulate incoming light so that the 4D light field can be computationally recovered from a single 2D photograph without extra refractive elements; a mask at the aperture instead allows full-resolution refocusing of layered scenes.[30]
  • Synthetic light fields: In computer-generated imagery, light fields can be generated synthetically by rendering a scene from multiple viewpoints; Levoy and Hanrahan built light fields from both rendered and photographed images.[16] This is relevant for VR and AR, where scenes might be entirely or partially virtual.
  • Microscopy: Placing a microlens array at the intermediate image plane of a microscope captures a light field of a specimen in a single photograph, at the cost of spatial resolution.[1]

Light field processing

Raw light field data requires significant processing:

  • Calibration and rectification: Raw data needs calibration to correct for lens distortions and misalignments, especially in camera arrays or plenoptic systems. Geometric calibration of microlens-based cameras, for example from line features in the raw images, is a research topic of its own.[31]
  • Interpolation and view synthesis: A key advantage is generating novel viewpoints not explicitly captured. Kalantari, Wang and Ramamoorthi used two convolutional neural networks to synthesize new views from only the four corner sub-aperture views of a Lytro Illum capture, with the aim of lowering the angular resolution a camera needs.[29]
  • Depth estimation: The angular variation of light rays encodes depth. Because a single light field capture contains both defocus and correspondence cues, Tao and colleagues combined the two into dense depth maps usable for matting, depth-of-field control and surface reconstruction.[32]
  • Compression: Light field datasets are large. A 1999 Stanford light field of Michelangelo's Night in the Medici Chapel contained 24,000 images of 1.3 megapixels each, and Lytro's 2016 Cinema camera could produce up to 400 gigabytes of data per second.[1][33] Researchers have compared and evaluated many light field image coding approaches.[34] For Welcome to Light Fields, Google modified the VP9 video codec to give high-quality compression with real-time, random-access decompression.[35][36]
  • Standards: JPEG Pleno Part 2, "Light field coding" (ISO/IEC 21794-2), was published in 2021, followed the same year by an amendment on profiles and levels. As of October 2026 JPEG lists a second edition of Part 2 at final draft stage and a new Part 7 on light field quality assessment at draft stage.[27] The MPEG Immersive Video (MIV) standard, ISO/IEC 23090-12, codes multi-camera content for playback with six degrees of freedom within a limited range of viewing positions. Its first edition was published on 11 August 2023 and was replaced by a 2025 edition.[37][38]

Light field rendering and display

Once captured or generated, light fields enable novel rendering and display possibilities.

Rendering

Rendering involves sampling the 4D light field data to synthesize a 2D image for a virtual camera. For a desired camera pose, the algorithm selects the rays that pass through the virtual camera and resamples their values from the stored data, possibly interpolating between nearby samples.[1] As long as the viewer stays outside the convex hull of the captured object, new perspectively correct views can be built from positions where no camera stood.[1] This enables:

  • Refocusing: Synthetically changing the focal plane after capture.[5][19]
  • Depth of field control: Computationally changing the effective aperture; refocusing with a plenoptic camera also extends depth of field without stopping down the lens, allowing shorter exposures.[5]
  • View synthesis: Generating views from slightly different positions, enabling parallax effects and six degrees of freedom (6DoF) movement within the captured volume. Google's light field app supports 6DoF viewing within a volume roughly 60 cm across.[36]
  • Neural Radiance Fields (NeRF) and neural rendering: NeRF represents a scene as a continuous function learned by a fully connected neural network that maps a 5D coordinate (3D location and 2D viewing direction) to volume density and view-dependent radiance; images are formed by volume rendering, and the network is trained from photographs with known camera poses.[25][39] Meta's VR-NeRF applied the approach to walkable spaces captured with a multi-camera rig, rendering at dual 2K x 2K resolution at 36 Hz on the authors' demo machine.[40] 3D Gaussian Splatting, introduced in 2023, represents a radiance field with 3D Gaussians and reports real-time novel-view synthesis at 100 frames per second or more at 1080p.[41]

Light field displays

Main article: Light field display

Light field displays aim to reproduce the directional aspect of light rays, allowing the viewer's eyes to accommodate (focus) at different depths, which can address the vergence-accommodation conflict of conventional stereoscopic displays.[23][6] In a conventional stereoscopic display, focus cues (accommodation and retinal blur) specify the depth of the screen rather than the depicted scene. Hoffman and colleagues found in 2008 that when focus cues were correct or nearly correct, viewers identified stereoscopic stimuli faster, perceived depth with less distortion, and reported less fatigue and discomfort.[42] Levoy noted in 2006 that displaying light fields is fundamentally harder than capturing them, because interpolation between sparse samples cannot be done digitally, only optically.[1] A 2021 review by Xiong and colleagues groups the main approaches to the accommodation cue in AR and VR displays as integral imaging, multifocal or varifocal displays, holographic displays and Maxwellian-view displays.[7] Key approaches include:

Integral imaging displays

These use a microlens array placed over a high-resolution display panel (for example OLED or LCD). Each microlens projects the pixels underneath it into different directions, creating multiple views of the scene. Densely sampled views approximate a continuous light field, enabling autostereoscopic viewing.[11] Lanman and Luebke describe integral imaging displays and light field cameras as sharing the same underlying concepts.[6] In near-eye form, Lanman and Luebke's 2013 microlens-based display synthesized light fields for virtual objects within the viewer's accommodation range and set out the trade-offs among resolution, field of view and depth of field.[6] Xiong et al. note that integral imaging has an inherent trade-off between the number of views and resolution, which makes it less attractive for VR headsets aiming at human visual acuity.[7]

Multi-plane and varifocal displays

These displays present images at multiple discrete focal planes, either simultaneously or in rapid sequence, or move a single focal plane to the depth the user is looking at, using eye tracking to find that depth. Varifocal displays show one image per frame and need eye tracking; multifocal displays present several images at different depths.[7] In a multifocal head-mounted display, several image planes span the viewer's accommodation range, and a decomposition algorithm distributes the scene across them; Mercier and colleagues built a vergence- and accommodation-tracked multifocal testbed with fast optimal decompositions of this kind.[43] Such displays approximate focus cues with a small number of image planes rather than reproducing a continuous light field.[43] See Varifocal display.

Holographic displays

Holographic displays reconstruct the light wavefront itself using spatial light modulators (SLMs) to control the phase or amplitude of light, producing a natural focus-and-blur effect like a real 3D object.[7] Li and colleagues demonstrated a see-through AR holographic display in 2016 in which a single holographic optical element acts as both mirror and lens.[44] NVIDIA and Stanford's Holographic Glasses for VR (2022) used a phase-only SLM in a binocular prototype 2.5 mm thick and weighing 60 g excluding the driving board, with a 22.8 degree diagonal field of view and a 2.3 mm static eyebox.[45][46] A 2024 follow-up in Nature combined inverse-designed metasurface gratings, a compact waveguide and AI-driven holography for full-color 3D holographic AR.[47][46]

Compressive and tensor displays

These use multiple layers of modulating panels (for example LCDs) with computational algorithms to sculpt the light passing through them, synthesizing a target light field with relatively thin hardware. Layered 3D (2011) used tomographic techniques to compute patterns for stacks of light-attenuating layers, and tensor displays (2012) generalized the idea to time-multiplexed layers with uniform or directional backlighting, using nonnegative tensor factorization.[48][49] The Light field stereoscope applied factored light field synthesis to a VR headset with two stacked transparent LCD panels; because the eye box is only slightly larger than the pupil, rank-1 factorizations can produce correct or nearly correct focus cues.[23]

Projector and pinlight arrays

Systems using high-speed projectors directed onto specialized screens can also generate light fields. A 2007 display described by Jones and colleagues used a high-speed video projector and a spinning mirror covered by a holographic diffuser to show interactive 3D graphics to viewers all around it, projecting 360-degree views with 1.25 degree separation.[50] For near-eye AR, pinlight displays place an array of out-of-focus point light sources and an LCD directly in front of the eye to form tiled miniature see-through projectors; a 2014 prototype offered a 110 degree diagonal field of view in the form factor of large glasses.[51]

Commercial examples and prototypes

Several companies are developing light field or related display technologies:

  • Looking Glass Factory: Produces glasses-free light field displays that, according to the company, project dozens of perspectives at once (up to 100, depending on the model) so that several people can view a 3D image without headsets or tracking.[52]
  • Avegant: In March 2017 announced a light field display for AR that it said could show objects at multiple focal planes simultaneously; CTO Edward Tang said digital light fields use discrete focal planes and that Avegant's display interpolates between them. Engadget's hands-on described the prototype as a rough engineering prototype connected by thick cables to a nearby PC.[53][54]
  • Magic Leap: Before its first headset shipped, Magic Leap's John Root named light field display technology as a key near-future technology for the field.[55] The shipping Magic Leap One used color-specific waveguides to place imagery on two fixed focal planes, which UploadVR, citing iFixit's teardown, classified as a fixed multifocal display rather than a full light field display (see below).[56]
  • Leia Inc.: Founded in 2014 as a spin-out of HP Labs.[57] Its founder David Fattal and colleagues at HP Labs published a multi-directional diffractive backlight in 2013 that produced a 64-view, full-parallax 3D image with a 90 degree view zone, aimed at thin mobile devices.[58]
  • CREAL: Swiss company based in Lausanne developing light field micro-displays for AR glasses and vision care, specifically targeting the focusing problem of stereoscopic AR.[59] Road to VR tried a bulky proof-of-concept AR headset at AWE 2023.[60] In July 2025 CREAL closed a US$8.9 million round led by ZEISS, bringing its total funding to over US$32 million, and said it would license its technology for ZEISS diagnostic and treatment devices while continuing to miniaturize its AR module.[59][61]
  • Light Field Lab: Based in San Jose, California; in October 2021 it unveiled SolidLight, a modular holographic display platform built from 28-inch panels that the company said each contribute 2.5 billion pixels.[62]

Terminology: what counts as a light field display

Whether a product is a "true" light field display is debated, because many devices sample the light field coarsely or provide only a few focal planes. Avegant's own CTO described digital light fields as built from discrete focal planes.[53] UploadVR wrote that the Magic Leap One was "not exactly a light field display" after Magic Leap had raised over US$2 billion while promoting a photonic light field chip, since its waveguides provide two fixed focal planes.[56] Google described the display of its Project Starline telepresence booth as a "light field display system", while the system's technical paper describes a head-tracked autostereoscopic display and a lenticular display among its components.[63][64] The 2021 review by Xiong and colleagues treats multifocal and varifocal displays, holographic displays and integral imaging displays as separate approaches to producing the accommodation cue.[7]

Applications in VR and AR

Light field technology has several applications in VR and AR:

  • Addressing the vergence-accommodation conflict (VAC): By providing focus cues that match the vergence depth, light field displays aim to reduce the eye strain and visual fatigue associated with conventional stereoscopic HMDs; laboratory work has shown that correct focus cues reduce fatigue and discomfort.[42][60] Road to VR's 2023 hands-on noted that a light field display can also apply a digital vision prescription to virtual content, and that the added immersion probably mattered mainly for objects within arm's reach.[60]
  • Correct parallax and 6DoF: Light fields support correct parallax shifts as the user moves the head within the captured volume, which matters for 6DoF experiences in VR and stable anchoring of virtual objects in AR; new, perspective-correct views can be built for positions where no camera stood, as long as the viewer stays outside the convex hull of the captured objects.[1] Google's Welcome to Light Fields VR app demonstrated this effect; released free on SteamVR in March 2018 for the HTC Vive, Oculus Rift and Windows Mixed Reality headsets, it let users lean within a volume about two feet wide.[4] Its rendering pipeline generated stereo views at 90 Hz on commodity VR hardware.[35]
  • Enhanced visual fidelity and view-dependent effects: Light fields reproduce light interactions such as specular highlights and reflections, which change with viewing direction; Google's light field demo showed reflections and lighting changing correctly as the head moved.[4] Neural representations such as NeRF model this view-dependent radiance explicitly.[25]
  • Realistic capture for VR content: Light field cameras capture real scenes with more information than 360-degree video, preserving view-dependent lighting and allowing natural head movement in VR. Lytro Immerge, announced in November 2015, used hundreds of cameras arranged in rings to record 360-degree light field video that viewers could explore within a volume of about one meter.[24] Google's DeepView research system (2020) captured light field video with a 46-camera dome and converted it into a layered mesh representation that could be played back on PC VR headsets, mobile VR and AR platforms, or in a web browser.[26]
  • Light field passthrough for mixed reality: In VR video passthrough, cameras cannot sit where the eyes are, so they capture a different perspective than the user would see without the headset; current headsets correct this in software through reprojection, which can introduce distortion.[28][65] Meta Reality Labs' Flamera prototype, shown at SIGGRAPH 2023, instead uses a grid of lenses with an aperture behind each one that passes only the rays that would have reached the eye; the raw image is rearranged and completed with a depth map.[66][67] IEEE Spectrum reported that the prototype won Best in Show at SIGGRAPH 2023, while noting its limited field of view, a depth of field that is not continuous, and that it is a research prototype not close to shipping.[65]
  • Telepresence and remote collaboration: Google's Project Starline, announced in May 2021, showed a remote participant life-size and in three dimensions without glasses or headsets.[63] In a study reported in its technical paper, participants rated it above 2D videoconferencing on measures such as presence and attentiveness.[64] In May 2025 Google renamed the effort Google Beam, pairing an AI volumetric video model with a light field display, and said the first HP devices would reach select customers later that year.[68] In September 2026 Computerworld reported that Beam had become generally available for purchase, built into HP's Dimension video meeting hardware (HP Dimension with Google Beam), initially in Canada, France, Germany, Japan, the UK and the US.[69]

Advantages

  • Provides realistic views with correct parallax and view-dependent effects.[1]
  • Enables post-capture refocusing and depth of field adjustments (primarily a capture advantage).[5]
  • Can reduce the vergence-accommodation conflict in HMDs when focus cues are correct or nearly correct.[42]
  • Captures scene information useful for computational photography and computer vision tasks, such as depth estimation.[32]
  • Enables perspective-correct integration of the real world in passthrough mixed reality, as in the Flamera prototype.[28]

Challenges and limitations

  • Data size and management: 4D light fields contain far more data than conventional images, which complicates capture, storage, streaming and processing; a single light field of Michelangelo's Night contained 24,000 images, and Lytro Cinema could generate up to 400 GB of data per second.[1][33] Lytro Immerge was tethered to a dedicated server for recording because of its raw data volume.[24]
  • Computational complexity: Rendering for a 4D light field display costs more than rendering 2D images. Sun and colleagues addressed this with foveated light field rendering that, in their prototype, rendered only 16 to 30 percent of the rays while preserving perceived quality.[70]
  • Capture hardware complexity and cost: High-quality capture systems (plenoptic cameras, large camera arrays) remain complex and expensive; Google built two custom rigs for its panoramic light field stills, and Lytro rented its Cinema system in packages starting at US$125,000 rather than selling it.[35][33]
  • Display technology trade-offs: Light field displays suitable for consumer VR and AR HMDs (high resolution, high brightness, wide field of view, large eye box, low latency, compact form factor) are still largely under development. Current technologies involve trade-offs, for example among resolution, field of view and depth of field in microlens near-eye displays, and between view count and resolution in integral imaging.[6][7]
  • Limited angular resolution: Practical systems often have limited angular resolution, which can restrict the range of parallax and how fully they resolve the VAC; in a plenoptic camera, refocusing sharpness grows with directional resolution, which competes with spatial resolution for the same sensor pixels.[19][1]
  • Eye-box size: Some approaches, especially holographic ones, have a small viewing zone where the effect is seen correctly; NVIDIA and Stanford's holographic glasses had a 2.3 mm static eyebox and used beam steering to provide a dynamic eyebox.[46][45]

Future directions

Research and development continue in several directions:

  • Neural radiance fields and neural rendering: Scene representations learned with machine learning, such as NeRF and 3D Gaussian Splatting represent and render complex scenes with view-dependent effects from ordinary photographs, and have been applied to walkable VR captures.[25][41][40]
  • Eye-tracked foveated light fields: Combining eye tracking with light field rendering allows concentrating detail and computation where the user is looking (foveated rendering), making real-time performance more feasible.[70]
  • Compact light field optics: Development of metalenses, diffractive optics, metasurface waveguides and holographic optical elements aims to create thinner, lighter optics for near-eye 3D displays suitable for glasses-like AR and VR devices.[47][46][44]
  • Light field video coding and streaming: Standards such as JPEG Pleno for light field images and MPEG Immersive Video for multi-camera 6DoF content provide common formats for storing and transmitting this data, with further JPEG Pleno parts in development as of 2026.[27][37]
  • Commercial near-eye light field modules: Companies such as CREAL are miniaturizing light field display modules for AR glasses; as of its July 2025 funding announcement, CREAL was still advancing its module toward integration into lightweight glasses for enterprise and, later, consumer markets.[59]

See also

References

  1. ↑ 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 1.10 1.11 1.12 1.13 1.14 1.15 1.16 1.17 1.18 1.19 1.20 1.21 1.22 1.23 1.24 1.25 1.26 1.27 1.28 1.29 1.30 1.31 1.32 1.33 1.34 Marc Levoy (2006-08). "Light Fields and Computational Imaging". Computer (IEEE Computer Society), vol. 39, no. 8, pp. 46-55. doi:10.1109/MC.2006.270. https://graphics.stanford.edu/papers/lfphoto/levoy-lfphoto-ieee06.pdf. Retrieved 2026-10-13.
  2. ↑ 2.0 2.1 2.2 2.3 Marc Levoy, Pat Hanrahan (1996-08). "Light field rendering". Proceedings of the 23rd Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH '96), pp. 31-42. ACM. https://doi.org/10.1145/237170.237199. Retrieved 2026-10-13.
  3. ↑ 3.0 3.1 3.2 3.3 Steven J. Gortler, Radek Grzeszczuk, Richard Szeliski, Michael F. Cohen (1996-08). "The Lumigraph". Proceedings of the 23rd Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH '96), pp. 43-54. ACM. https://doi.org/10.1145/237170.237200. Retrieved 2026-10-13.
  4. ↑ 4.0 4.1 4.2 4.3 4.4 Paul Debevec (2018-03-14). "Experimenting with Light Fields". The Keyword (Google blog). https://blog.google/products-and-platforms/products/google-ar-vr/experimenting-light-fields/. Retrieved 2026-10-13.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 5.5 5.6 5.7 Ren Ng, Marc Levoy, Mathieu Bredif, Gene Duval, Mark Horowitz, Pat Hanrahan (2005-04). "Light Field Photography with a Hand-Held Plenoptic Camera". Stanford University Computer Science Tech Report CSTR 2005-02. https://graphics.stanford.edu/papers/lfcamera/. Retrieved 2026-10-13.
  6. ↑ 6.0 6.1 6.2 6.3 6.4 6.5 Douglas Lanman, David Luebke (2013-11). "Near-eye light field displays". ACM Transactions on Graphics, vol. 32, no. 6 (SIGGRAPH Asia 2013). ACM. https://doi.org/10.1145/2508363.2508366. Retrieved 2026-10-13.
  7. ↑ 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7 Jianghao Xiong, En-Lin Hsiang, Ziqian He, Tao Zhan, Shin-Tson Wu (2021-10-25). "Augmented reality and virtual reality displays: emerging technologies and future perspectives". Light: Science & Applications, vol. 10. https://doi.org/10.1038/s41377-021-00658-8. Retrieved 2026-10-13.
  8. ↑ 8.0 8.1 Michael Faraday (1846-05). "Thoughts on Ray-Vibrations". The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, vol. 28, no. 188, pp. 345-350. https://doi.org/10.1080/14786444608645431. Retrieved 2026-10-13.
  9. ↑ 9.0 9.1 A. Gershun (1939-04). "The Light Field". Journal of Mathematics and Physics, vol. 18, pp. 51-151. https://doi.org/10.1002/sapm193918151. Retrieved 2026-10-13.
  10. ↑ 10.0 10.1 G. Lippmann (1908). "Épreuves réversibles donnant la sensation du relief". Journal de Physique Théorique et Appliquée, vol. 7, pp. 821-825. https://doi.org/10.1051/jphystap:019080070082100. Retrieved 2026-10-13.
  11. ↑ 11.0 11.1 Manuel Martínez-Corral, Bahram Javidi (2018-07-03). "Fundamentals of 3D imaging and displays: a tutorial on integral imaging, light-field, and plenoptic systems". Advances in Optics and Photonics, vol. 10, no. 3, p. 512. Optica Publishing Group. https://doi.org/10.1364/AOP.10.000512. Retrieved 2026-10-13.
  12. ↑ 12.0 12.1 12.2 12.3 12.4 12.5 Edward H. Adelson, James R. Bergen (1991). "The Plenoptic Function and the Elements of Early Vision". Computational Models of Visual Processing (M. Landy and J. A. Movshon, eds.), pp. 3-20. MIT Press. https://www.cns.nyu.edu/~tony/vns/readings/adelson-bergen-1991.pdf. Retrieved 2026-10-13.
  13. ↑ 13.0 13.1 13.2 E. H. Adelson, J. Y. A. Wang (1992). "Single lens stereo with a plenoptic camera". IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 14, no. 2, pp. 99-106. https://doi.org/10.1109/34.121783. Retrieved 2026-10-13.
  14. ↑ Leonard McMillan, Gary Bishop (1995). "Plenoptic modeling: an image-based rendering system". Proceedings of the 22nd Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH '95), pp. 39-46. ACM. https://doi.org/10.1145/218380.218398. Retrieved 2026-10-13.
  15. ↑ 15.0 15.1 "Light fields and computational photography". Stanford Computer Graphics Laboratory. https://graphics.stanford.edu/projects/lightfield/. Retrieved 2026-10-13.
  16. ↑ 16.0 16.1 16.2 16.3 16.4 16.5 16.6 Marc Levoy, Pat Hanrahan. "Light Field Rendering (project page)". Stanford Computer Graphics Laboratory. https://graphics.stanford.edu/papers/light/. Retrieved 2026-10-13.
  17. ↑ Jin-Xiang Chai, Shing-Chow Chan, Heung-Yeung Shum, Xin Tong (2000). "Plenoptic sampling". Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH 2000), pp. 307-318. ACM. https://doi.org/10.1145/344779.344932. Retrieved 2026-10-13.
  18. ↑ 18.0 18.1 18.2 Bennett Wilburn, Neel Joshi, Vaibhav Vaish, Eino-Ville Talvala, Emilio Antunez, Adam Barth, Andrew Adams, Mark Horowitz, Marc Levoy (2005-07). "High performance imaging using large camera arrays". ACM Transactions on Graphics, vol. 24, no. 3, pp. 765-776 (SIGGRAPH 2005). doi:10.1145/1073204.1073259. https://graphics.stanford.edu/papers/CameraArray/. Retrieved 2026-10-13.
  19. ↑ 19.0 19.1 19.2 Ren Ng (2005-07). "Fourier slice photography". ACM Transactions on Graphics, vol. 24, no. 3, pp. 735-744 (SIGGRAPH 2005). https://doi.org/10.1145/1073204.1073256. Retrieved 2026-10-13.
  20. ↑ 20.0 20.1 Dante Cesa (2011-10-19). "Lytro introduces world's first light field camera: f/2 lens, $399, ships early 2012". Engadget. https://www.engadget.com/2011-10-19-lytro-introduces-worlds-first-lightfield-camera.html. Retrieved 2026-10-13.
  21. ↑ 21.0 21.1 Megan Rose Dickey (2018-03-27). "Lytro, a light-field imaging startup, is shutting down". TechCrunch. https://techcrunch.com/2018/03/27/lytro-is-shutting-down/. Retrieved 2026-10-13.
  22. ↑ 22.0 22.1 "Technology". Raytrix. https://www.raytrix.de/technology/. Retrieved 2026-10-13.
  23. ↑ 23.0 23.1 23.2 Fu-Chung Huang, Kevin Chen, Gordon Wetzstein (2015-07). "The Light Field Stereoscope: Immersive Computer Graphics via Factored Near-Eye Light Field Displays with Focus Cues". ACM Transactions on Graphics, vol. 34, no. 4 (SIGGRAPH 2015). doi:10.1145/2766922. https://www.computationalimaging.org/publications/the-light-field-stereoscope/. Retrieved 2026-10-13.
  24. ↑ 24.0 24.1 24.2 Will Mason (2015-11-04). "Lytro announces world's first light field VR video camera". UploadVR. https://www.uploadvr.com/lytro-immerge-vr-light-field-video-camera/. Retrieved 2026-10-13.
  25. ↑ 25.0 25.1 25.2 25.3 Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, Ren Ng (2020). "NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis". European Conference on Computer Vision (ECCV 2020), Lecture Notes in Computer Science, pp. 405-421. Springer. https://doi.org/10.1007/978-3-030-58452-8_24. Retrieved 2026-10-13.
  26. ↑ 26.0 26.1 26.2 Michael Broxton, John Flynn, Ryan Overbeck, Daniel Erickson, Peter Hedman, Matthew DuVall, Jason Dourgarian, Jay Busch, Matt Whalen, Paul Debevec (2020). "Immersive Light Field Video with a Layered Mesh Representation". ACM Transactions on Graphics, vol. 39, no. 4 (SIGGRAPH 2020). Google. doi:10.1145/3386569.3392485. https://augmentedperception.github.io/deepviewvideo/. Retrieved 2026-10-13.
  27. ↑ 27.0 27.1 27.2 "JPEG Pleno work plan". JPEG (ISO/IEC JTC 1/SC 29/WG 1). https://jpeg.org/jpegpleno/workplan.html. Retrieved 2026-10-13.
  28. ↑ 28.0 28.1 28.2 Grace Kuo, Eric Penner, Seth Moczydlowski, Alexander Ching, Douglas Lanman, Nathan Matsuda (2023-07-23). "Perspective-Correct VR Passthrough Without Reprojection". SIGGRAPH 2023 Conference Proceedings, pp. 1-9. ACM. https://doi.org/10.1145/3588432.3591534. Retrieved 2026-10-13.
  29. ↑ 29.0 29.1 Nima Khademi Kalantari, Ting-Chun Wang, Ravi Ramamoorthi (2016-11). "Learning-based view synthesis for light field cameras". ACM Transactions on Graphics, vol. 35, no. 6 (SIGGRAPH Asia 2016). https://doi.org/10.1145/2980179.2980251. Retrieved 2026-10-13.
  30. ↑ Ashok Veeraraghavan, Ramesh Raskar, Amit Agrawal, Ankit Mohan, Jack Tumblin (2007-07). "Dappled photography: mask enhanced cameras for heterodyned light fields and coded aperture refocusing". ACM Transactions on Graphics, vol. 26, no. 3 (SIGGRAPH 2007). https://doi.org/10.1145/1276377.1276463. Retrieved 2026-10-13.
  31. ↑ Yunsu Bok, Hae-Gon Jeon, In So Kweon (2017-02). "Geometric Calibration of Micro-Lens-Based Light Field Cameras Using Line Features". IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 2, pp. 287-300. https://doi.org/10.1109/TPAMI.2016.2541145. Retrieved 2026-10-13.
  32. ↑ 32.0 32.1 Michael W. Tao, Sunil Hadap, Jitendra Malik, Ravi Ramamoorthi (2013-12). "Depth from Combining Defocus and Correspondence Using Light-Field Cameras". 2013 IEEE International Conference on Computer Vision (ICCV), pp. 673-680. https://doi.org/10.1109/ICCV.2013.89. Retrieved 2026-10-13.
  33. ↑ 33.0 33.1 33.2 Lucas Matney (2016-04-11). "Lytro's 755 megapixel Cinema light field camera is going to kill the green screen". TechCrunch. https://techcrunch.com/2016/04/11/lytro-cinema-is-giving-filmmakers-400-gigabytes-per-second-of-creative-freedom. Retrieved 2026-10-13.
  34. ↑ Irene Viola, Martin Rerabek, Touradj Ebrahimi (2017-10). "Comparison and Evaluation of Light Field Image Coding Approaches". IEEE Journal of Selected Topics in Signal Processing, vol. 11, no. 7, pp. 1092-1106. https://doi.org/10.1109/JSTSP.2017.2740167. Retrieved 2026-10-13.
  35. ↑ 35.0 35.1 35.2 Ryan S. Overbeck, Daniel Erickson, Daniel Evangelakos, Matt Pharr, Paul Debevec (2018-12). "A system for acquiring, processing, and rendering panoramic light field stills for virtual reality". ACM Transactions on Graphics, vol. 37, no. 6 (SIGGRAPH Asia 2018). https://doi.org/10.1145/3272127.3275031. Retrieved 2026-10-13.
  36. ↑ 36.0 36.1 "Welcome to Light Fields: A System for Acquiring, Processing, and Rendering Panoramic Light Field Stills for Virtual Reality". Google Augmented Perception. https://augmentedperception.github.io/welcome-to-lightfields/. Retrieved 2026-10-13.
  37. ↑ 37.0 37.1 "ISO/IEC 23090-12:2023 Information technology - Coded representation of immersive media - Part 12: MPEG immersive video". Standard Norge. https://online.standard.no/en/isoiec-23090-12-2023. Retrieved 2026-10-13.
  38. ↑ Jill M. Boyce, Renaud Dore, Adrian Dziembowski, Julien Fleureau, Joel Jung, Bart Kroon, Basel Salahieh, Vinod Kumar Malamal Vadakital, Lu Yu (2021-09). "MPEG Immersive Video Coding Standard". Proceedings of the IEEE, vol. 109, no. 9, pp. 1521-1536. https://doi.org/10.1109/JPROC.2021.3062590. Retrieved 2026-10-13.
  39. ↑ Ben Mildenhall et al.. "NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis (project page)". Matthew Tancik. https://www.matthewtancik.com/nerf. Retrieved 2026-10-13.
  40. ↑ 40.0 40.1 Linning Xu et al. (2023-12). "VR-NeRF: High-Fidelity Virtualized Walkable Spaces". SIGGRAPH Asia 2023 Conference Proceedings. Meta, The Chinese University of Hong Kong. doi:10.1145/3610548.3618139. https://vr-nerf.github.io/. Retrieved 2026-10-13.
  41. ↑ 41.0 41.1 Bernhard Kerbl, Georgios Kopanas, Thomas Leimkuehler, George Drettakis (2023-07). "3D Gaussian Splatting for Real-Time Radiance Field Rendering". ACM Transactions on Graphics, vol. 42, no. 4 (SIGGRAPH 2023). Inria. doi:10.1145/3592433. https://repo-sam.inria.fr/fungraph/3d-gaussian-splatting/. Retrieved 2026-10-13.
  42. ↑ 42.0 42.1 42.2 David M. Hoffman, Ahna R. Girshick, Kurt Akeley, Martin S. Banks (2008-03-28). "Vergence-accommodation conflicts hinder visual performance and cause visual fatigue". Journal of Vision, vol. 8, no. 3, article 33. https://doi.org/10.1167/8.3.33. Retrieved 2026-10-13.
  43. ↑ 43.0 43.1 Olivier Mercier, Yusufu Sulai, Kevin Mackenzie, Marina Zannoli, James Hillis, Derek Nowrouzezahrai, Douglas Lanman (2017-11). "Fast gaze-contingent optimal decompositions for multifocal displays". ACM Transactions on Graphics, vol. 36, no. 6 (SIGGRAPH Asia 2017). https://doi.org/10.1145/3130800.3130846. Retrieved 2026-10-13.
  44. ↑ 44.0 44.1 Gang Li, Dukho Lee, Youngmo Jeong, Jaebum Cho, Byoungho Lee (2016). "Holographic display for see-through augmented reality using mirror-lens holographic optical element". Optics Letters, vol. 41, no. 11, pp. 2486-2489. https://doi.org/10.1364/OL.41.002486. Retrieved 2026-10-13.
  45. ↑ 45.0 45.1 Jonghyun Kim, Manu Gopakumar, Suyeon Choi, Yifan Peng, Ward Lopes, Gordon Wetzstein (2022-08). "Holographic Glasses for Virtual Reality". SIGGRAPH 2022 Conference Proceedings, pp. 1-9. ACM. https://doi.org/10.1145/3528233.3530739. Retrieved 2026-10-13.
  46. ↑ 46.0 46.1 46.2 46.3 Jonghyun Kim (2024-06-14). "Developing Smaller, Lighter Extended Reality Glasses Using AI". NVIDIA Technical Blog. https://developer.nvidia.com/blog/developing-smaller-lighter-extended-reality-glasses-using-ai/. Retrieved 2026-10-13.
  47. ↑ 47.0 47.1 Manu Gopakumar, Gun-Yeal Lee, Suyeon Choi, Brian Chao, Yifan Peng, Jonghyun Kim, Gordon Wetzstein (2024-05-08). "Full-colour 3D holographic augmented-reality displays with metasurface waveguides". Nature, vol. 629, pp. 791-797. https://doi.org/10.1038/s41586-024-07386-0. Retrieved 2026-10-13.
  48. ↑ Gordon Wetzstein, Douglas Lanman, Wolfgang Heidrich, Ramesh Raskar (2011-07). "Layered 3D: tomographic image synthesis for attenuation-based light field and high dynamic range displays". ACM Transactions on Graphics, vol. 30, no. 4 (SIGGRAPH 2011). https://doi.org/10.1145/2010324.1964990. Retrieved 2026-10-13.
  49. ↑ Gordon Wetzstein, Douglas Lanman, Matthew Hirsch, Ramesh Raskar (2012-07). "Tensor displays: compressive light field synthesis using multilayer displays with directional backlighting". ACM Transactions on Graphics, vol. 31, no. 4 (SIGGRAPH 2012). https://doi.org/10.1145/2185520.2185576. Retrieved 2026-10-13.
  50. ↑ Andrew Jones, Ian McDowall, Hideshi Yamada, Mark Bolas, Paul Debevec (2007-07). "Rendering for an interactive 360° light field display". ACM Transactions on Graphics, vol. 26, no. 3 (SIGGRAPH 2007). https://doi.org/10.1145/1276377.1276427. Retrieved 2026-10-13.
  51. ↑ Andrew Maimone, Douglas Lanman, Kishore Rathinavel, Kurtis Keller, David Luebke, Henry Fuchs (2014-07). "Pinlight displays: wide field of view augmented reality eyeglasses using defocused point light sources". ACM Transactions on Graphics, vol. 33, no. 4 (SIGGRAPH 2014). https://doi.org/10.1145/2601097.2601141. Retrieved 2026-10-13.
  52. ↑ "Looking Glass". Looking Glass Factory. https://lookingglassfactory.com/. Retrieved 2026-10-13.
  53. ↑ 53.0 53.1 Ben Lang (2017-03-21). "Avegant Claims Newly Announced Display Tech is "a new method to create light fields"". Road to VR. https://roadtovr.com/avegants-newly-announced-light-field-tech-claims-new-method-display-light-fields/. Retrieved 2026-10-13.
  54. ↑ Nicole Lee (2017-05-01). "Avegant's light-field tech gives hope to a mixed-reality future". Engadget. https://www.engadget.com/2017-05-01-avegant-light-field-hands-on.html. Retrieved 2026-10-13.
  55. ↑ Mike Seymour (2015-05-26). "Light fields - the future of VR-AR-MR". fxguide. https://www.fxguide.com/fxfeatured/light-fields-the-future-of-vr-ar-mr/. Retrieved 2026-10-13.
  56. ↑ 56.0 56.1 Ian Hamilton (2018-08-27). "What Magic Leap One And Facebook's Half Dome Have In Common". UploadVR. https://uploadvr.com/magic-leap-one-half-dome-common-differences. Retrieved 2026-10-13.
  57. ↑ "Leia Inc. Strengthens Board with Global Technology Leaders to Help Scale the Next Era of Spatial AI". Immersity (Leia Inc.). 2026-07-29. https://immersity.ai/news-articles/leia-inc-strengthens-board-with-global-technology-leaders-to-help-scale-the-next-era-of-spatial-ai. Retrieved 2026-10-13.
  58. ↑ David Fattal, Zhen Peng, Tho Tran, Sonny Vo, Marco Fiorentino, Jim Brug, Raymond G. Beausoleil (2013-03-21). "A multi-directional backlight for a wide-angle, glasses-free three-dimensional display". Nature, vol. 495, no. 7441, pp. 348-351. https://doi.org/10.1038/nature11972. Retrieved 2026-10-13.
  59. ↑ 59.0 59.1 59.2 "CREAL Closes $8.9M Investment Round Led by ZEISS to Advance Vision Care and AR". CREAL. 2025-07-10. https://creal.com/app/uploads/2025/07/2025-07-10-CREAL-Investment-Round-Press-Release.pdf. Retrieved 2026-10-13.
  60. ↑ 60.0 60.1 60.2 Ben Lang (2023-06-03). "Hands-on: CREAL's Light-field Display Brings a New Layer of Immersion to AR". Road to VR. https://www.roadtovr.com/creal-light-field-display-new-immersion-ar/. Retrieved 2026-10-13.
  61. ↑ "CREAL: Light-field Display Technology". CREAL. https://creal.com/. Retrieved 2026-10-13.
  62. ↑ "Light Field Lab Unveils SolidLight, the Highest Resolution Holographic Display Platform Ever Designed". PR Newswire. Light Field Lab. 2021-10-07. https://www.prnewswire.com/news-releases/light-field-lab-unveils-solidlight--the-highest-resolution-holographic-display-platform-ever-designed-301395002.html. Retrieved 2026-10-13.
  63. ↑ 63.0 63.1 Clay Bavor (2021-05-18). "Project Starline: Feel like you're there, together". The Keyword (Google blog). https://blog.google/technology/research/project-starline/. Retrieved 2026-10-13.
  64. ↑ 64.0 64.1 Jason Lawrence et al. (2021-12). "Project Starline: a high-fidelity telepresence system". ACM Transactions on Graphics, vol. 40, no. 6. https://doi.org/10.1145/3478513.3480490. Retrieved 2026-10-13.
  65. ↑ 65.0 65.1 Matthew S. Smith (2023-08-25). "Meta's Flamera Has a New Vision for Augmented Reality". IEEE Spectrum. https://spectrum.ieee.org/meta-flamera. Retrieved 2026-10-13.
  66. ↑ Jan Wöbbeking (2023-08-01). "Meta's Flamera headset prototype fights distortion with "bug eyes"". MIXED. https://mixed-news.com/en/meta-shows-flamera-passthrough-ar-headset-prototype-at-siggraph/. Retrieved 2026-10-13.
  67. ↑ Shawnee Blackwood (2023-08-04). "Meta's Going to Siggraph 2023 and Showing Flamera and Butterscotch VR Technologies". Display Daily. https://displaydaily.com/metas-going-to-siggraph-2023-and-showing-flamera-and-butterscotch-vr-technologies/. Retrieved 2026-10-13.
  68. ↑ Andrew Nartker (2025-05-20). "Google Beam: Our AI-first 3D video communication platform". The Keyword (Google blog). https://blog.google/technology/research/project-starline-google-beam-update/. Retrieved 2026-10-13.
  69. ↑ Matthew Finnegan (2026-09-24). "Five years later, you can finally buy Google Beam". Computerworld. https://computerworld.com/article/4225627/five-years-later-you-can-finally-buy-google-beam.html. Retrieved 2026-10-13.
  70. ↑ 70.0 70.1 Qi Sun, Fu-Chung Huang, Joohwan Kim, Li-Yi Wei, David Luebke, Arie Kaufman (2017-11-27). "Perceptually-Guided Foveation for Light Field Displays". NVIDIA Research (ACM SIGGRAPH Asia 2017). doi:10.1145/3130800.3130807. https://research.nvidia.com/publication/2017-11_perceptually-guided-foveation-light-field-displays. Retrieved 2026-10-13.
Retrieved from "https://vrarwiki.com/index.php?title=Light_field&oldid=51751"
Contents