US20140327748A1 - Rendering for an interactive 360 degree light field display - Google Patents
Rendering for an interactive 360 degree light field display Download PDFInfo
- Publication number
- US20140327748A1 US20140327748A1 US13/850,229 US201313850229A US2014327748A1 US 20140327748 A1 US20140327748 A1 US 20140327748A1 US 201313850229 A US201313850229 A US 201313850229A US 2014327748 A1 US2014327748 A1 US 2014327748A1
- Authority
- US
- United States
- Prior art keywords
- mirror
- display
- viewer
- viewing plane
- projector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- H04N13/0402—
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/302—Image reproducers for viewing without the aid of special glasses, i.e. using autostereoscopic displays
-
- H04N13/0468—
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/302—Image reproducers for viewing without the aid of special glasses, i.e. using autostereoscopic displays
- H04N13/305—Image reproducers for viewing without the aid of special glasses, i.e. using autostereoscopic displays using lenticular lenses, e.g. arrangements of cylindrical lenses
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/324—Colour aspects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/349—Multi-view displays for displaying three or more geometrical viewpoints without viewer tracking
- H04N13/354—Multi-view displays for displaying three or more geometrical viewpoints without viewer tracking for displaying sequentially
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/363—Image reproducers using image projection screens
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/366—Image reproducers using viewer tracking
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/366—Image reproducers using viewer tracking
- H04N13/368—Image reproducers using viewer tracking for two or more viewers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/366—Image reproducers using viewer tracking
- H04N13/373—Image reproducers using viewer tracking for tracking forward-backward translational head movements, i.e. longitudinal movements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/366—Image reproducers using viewer tracking
- H04N13/38—Image reproducers using viewer tracking for tracking vertical translational head movements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/30—Image reproducers
- H04N13/366—Image reproducers using viewer tracking
- H04N13/376—Image reproducers using viewer tracking for tracking left-right translational head movements, i.e. lateral movements
Definitions
- An interactive, autostereoscopic system for displaying an object in three dimensions includes a mirror configured to spin around a vertical axis when actuated by a motor, a high speed video projector, and a processing system including a graphics card interfaced to the video projector.
- An anisotropic reflector is bonded onto an inclined surface of the mirror.
- the video projector projects video signals of the object from the projector onto the projection surface while the mirror is spinning, so that light rays representing the video signals are redirected toward a field of view of a 360 degree range within a viewing space of one or more viewers.
- the processing system renders the redirected light rays so as to generate a horizontal-parallax 3D display of the object for the viewers from any of a plurality of viewing directions within the viewing space.
- FIG. 1 illustrates a system for generating an autostereoscopic, multiview (horizontal-parallax-only) 3D display of an object, in one embodiment of the present disclosure.
- FIG. 2 illustrates the anisotropic reflectance characteristics of the mirror system illustrated in FIG. 1 .
- FIGS. 3A , 3 B, and 3 C illustrate a projection process and geometry for rendering an object or scene to the system illustrated in FIG. 1 , in one embodiment of the present disclosure.
- FIG. 4 illustrates a geometric calibration for the system illustrated in FIG. 1 , in one embodiment of the present disclosure.
- FIG. 5 illustrates a system for dynamic rendering of a 3D display from a photographically captured 4D light field data set, in one embodiment of the present disclosure.
- FIGS. 6A-6B illustrate dynamic vertical rebinning to generate correct vertical parallax for a horizontal-parallax-only 3D display of a photographically captured 4D light field data set, in one embodiment of the present disclosure.
- FIG. 7 illustrates a two-mirror tent for displaying two-toned color imagery using orange and cyan filters below the diffusers.
- the present disclosure describes an easily reproducible, low-cost 3D display system with a form factor for displaying 3D objects in 3D. Also described are the projection mathematics and rendering methods necessary to drive the 3D display with real time raster imagery or pre-recorded light fields so that they exhibit the correct cues of both horizontal and vertical parallax.
- the display is autostereoscopic, requiring no special viewing glasses, and omnidirectional, allowing viewers to be situated anywhere around it.
- the display is also multiview, i.e. produces a correct rendition of the light field with the correct horizontal parallax and vertical perspective for any viewpoint situated at a certain distance and height around the display.
- the projection equations described below allow the vertical parallax to be adjusted on the fly to allow the tracked users to properly see objects from arbitrary heights and distances, in addition to obtaining correct views from any angle around the display.
- the systems and methods described below use primarily commodity graphics and display components and achieve real-time rendering with non-trivial scene complexity across its entire field of view.
- FIG. 1 illustrates a system for generating an autostereoscopic 3D display of an object, in one embodiment of the present disclosure.
- the 3D display system 100 includes: a mirror 110 covered by an anisotropic reflector 112 , a motion-control motor 115 , a high-speed video projector 120 , and a processing system 130 (for example a standard PC) that includes a graphics card 132 interfaced to the video projector 120 using a custom FPGA-based image decoder.
- the mirror 110 is configured to spin with respect to a vertical axis, when actuated by the motion-control motor 115 .
- the spinning mirror 110 may be tilted or inclined at an angle with respect to the vertical axis, for example by about 45 degrees, to reflect rays of light from the projector 120 to all possible viewing positions around the device, allowing many viewers to view the display simultaneously.
- the video projector 120 achieves high-speed projection by modifying an off-the-shelf projector to use a new DLP drive card with custom programmed FPGA-based circuitry.
- the processing system 130 (containing the graphics card 132 ) continuously renders new horizontal views of the subject, which in the illustrated embodiment consists of 288 images per rotation of the mirror. These views are encoded into 24-bit images and sent to the video projector 120 .
- a complete kit consisting of the FPGA and DLP boards is now available from Polaris Road, Inc.
- the anisotropic reflector 112 is a holographic diffuser is bonded onto a surface of the mirror 110 .
- the anisotropic reflector 112 is a holographic diffuser.
- Other embodiments of the present disclosure may use different types of anisotropic reflectors, including but not limited to: a lenticular lens; and a lenslet array.
- the mirror surface (with the anisotropic reflector bonded thereon) provides a projection surface 114 , which reflects each projector pixel to a narrow range of viewpoints.
- the mirror surface (onto which the anisotropic reflector bonded) provides a projection surface that reflects incident rays of light in a way that the rays are spread in a substantially vertical direction after being reflected.
- This configuration contrasts with previous volumetric displays, which projected images onto a spinning diffuse plane which scattered light in all directions. Such displays could not recreate view-dependent effects such as occlusion.
- the holographic diffuser 112 provides control over the width and height of this region.
- the characteristics of the holographic diffuser 112 are such that the relative diffusion between x and y is approximately 1:200.
- the surface 114 is sharply specular to maintain a 1.25 degree separation between views.
- the mirror 110 scatters widely so the projected image can be viewed from essentially any height.
- FIG. 2 shows the anisotropic reflectance characteristics of the mirror system (i.e. the mirror 110 coupled to the motor 115 and having the holographic diffuser 112 bonded to a surface of the mirror 110 ).
- the horizontal profile of the specular lobe approximates a bilinear interpolation between adjacent viewpoints.
- the motion of the mirror may add some additional blur which improves reproduction of halftoned imagery at the expense of angular resolution.
- the holographic diffuser may be diffuse in the vertical dimension and sharply specular in the horizontal dimension.
- the dotted line plots the intensity of a laser beam as reflected by the holographic diffuser 112 and mirror 110 toward the viewers.
- the solid line illustrates the intensity of a thin vertical line of light from the video projector as reflected by the holographic diffuser 112 and mirror 110 toward the viewers.
- the dashed line illustrates the intensity of an ideal bilinear interpolation spread of a hat function whose radius matches the 1.25 degree angular separation of the successive views of the display system 100 .
- the anisotropic holographic diffuser 112 and mirror 110 may be mounted on a carbon fiber panel and attached to an aluminum fly-wheel at 45°, in the illustrated embodiment.
- the flywheel spins synchronously relative to the images displayed by the video projector 120 .
- the system 100 may be synchronized as follows. Since the output frame rate of the PC graphics card is relatively constant and cannot be fine-tuned on the fly, we use the PC video output rate as the master signal for system synchronization.
- the projector's FPGA also creates signals encoding the current frame rate.
- These control signals inter-face directly to an Animatics SM3420D “Smart Motor” which contains firmware and motion control parameters resulting in a stable, velocity-based control loop that ensures the motor velocity stays in sync with the signals from the projector. As the mirror rotates up to 20 times per second, persistence of vision creates the illusion of a floating object around the center of the mirror.
- the projector and spinning mirror yield a horizontal-parallax-only display; the image perspective does not change correctly as the viewpoint moves up and down, or forward and backward.
- the projection algorithms described further below can take into account the height and distance of the viewer to render the scene with correct perspective.
- the 3D, horizontal-parallax-only display may be augmented to include correct vertical parallax in the display.
- correct vertical parallax may be included by tracking the motion of one or more viewers, then interactively updating the displayed images of the object in response to the tracked motion of the one or more viewers.
- the displayed images of the object may be interactively updated in such a way that when the viewer is moving upward from below, the displayed image of the object appears to the viewer as if seen from above, and when the viewer is moving downward from above the displayed image of the object appears to the viewer as if seen from below.
- the displayed images of the object may be interactively updated in such a way that the images of the object as displayed to each viewer is shown from a viewpoint consistent with the height of each viewer relative to the display.
- a tracking system is thus used, for example where the viewer holds the tracking device (a sensor) to their temple, or to a video camera filming the display.
- the tracking system may be an electromagnetic motion tracking device, such as a Polhemus Patriot electromagnetic tracking device that can track the motion of a viewer when held by or attached to the viewer.
- the tracking system may include one or more video cameras.
- a plurality of video cameras may be used to observe a plurality of viewers, so that the position of each of the viewers can be determined.
- the tracking system may include a plurality of depth cameras used to track the position of a plurality of viewers.
- the tracking data may be used by the projection algorithm (further described below) to display the scene from the correct perspective for the viewer's height and distance.
- the display's horizontal parallax provides binocular stereo and yields zero lag as the viewer moves his/her head horizontally, which may be one of the most common significant head motions.
- the effects of vertical motion and distance change are computed based on the tracked position.
- the display only needs to adjust the rendered views in the vicinity of each tracked user, leaving the rest of the displayed circumference optimized to the average expected viewer position.
- FIGS. 3A , 3 B, and 3 C illustrate projection process and geometry for rendering an object and/or a scene to the system illustrated in FIG. 1 , in one embodiment of the present disclosure.
- the spinning mirror (labeled as 110 in FIG. 1 , re-labeled as 310 in FIG. 3A ) is centered at the origin and that its axis of rotation is the vertical y-axis, with the video projector at the nodal point P above the mirror, as shown in FIG. 3A .
- the viewpoint for which the correct perspective should be obtained is at a height h (relative to the center of the projection surface 114 ) and a distance d from the y- or vertical axis about which the mirror 310 rotates when actuated by the motion-control motor.
- perspective-correct imagery can be produced for any viewing position on a viewing space or viewing circle V defined by h and d. This yields binocular images for a viewer facing the display since h and d will be similar for both eyes.
- a particular viewpoint on the circle V is denoted in the present disclosure as V′.
- the set of perspective-correct viewpoints V need not be a continuous planar circle and can pass through a variety of tracked viewer positions at different distances and heights.
- each projector pixel (u, v) sends light toward one specific viewpoint V′ on V.
- the display system must ensure that each projected pixel displays the appropriate part of the scene as it should be seen from viewpoint V′.
- the scene being rendered is a polygonal 3D model, it must be determined, for any world-space vertex Q, where it should be rendered on the projector's image for any given mirror position.
- the 3D display system is considered as it would be viewed from above. It is noted that in the horizontal plane, the anisotropic mirror behaves in essence like a regular mirror. The optical path can thus be unfolded by reflecting the projector position P to P′ across the plane of the mirror 310 , as seen in FIG. 3B . A ray originating at P′ passing through Q will continue out into space toward the viewers. This ray will not, in general, intersect the view circle V.
- the vertical plane containing is intersected with the viewing circle V, to determine the viewpoint V′ from which Q will be seen with the mirror at its given position.
- this diffusion plane is an approximation to a cone-shaped reflection from the mirror, but the projection error is small for the setup described above, and may be neglected in practice.
- a ray is then traced from the viewpoint V′ toward Q, until it intersects the surface of the mirror at M.
- the symbol M represents the one point on the mirror that reflects light to the viewer coming from the direction of Q.
- M simply needs to be projected up toward the projector's nodal point P to find the corresponding projector pixel (u, v).
- illuminating a pixel at (u,v) will make it appear from viewpoint V′ that 3D point Q has been illuminated.
- Q will eventually be rendered as it should be seen from all other viewpoints on V as the mirror rotates.
- the technique actually renders multiple-center-of-projection (MCOP) images to the projector which can not be generated using a traditional projection matrix.
- MCOP multiple-center-of-projection
- the projection technique implemented by the processing system 130 uses a combination of two different viewpoints P (for horizontal coordinates) and V′ (for vertical coordinates). Nonetheless, the technique may easily be implemented as the vertex shader provided in Appendix B of the present disclosure, allowing an entire mesh to be rendered in a single pass. For z-buffering, vertex depth may be based on the distance from V′ to Q.
- projector pixel (u, v) reflects from mirror point M toward viewpoint V′.
- the color it should display should be the result of tracing a ray from V′ toward point M. If the scene being rendered is a light field, ray V′M is simply queried for the scene radiance at that point, as further described below.
- the fans of light from a given projector frame diverge horizontally toward multiple viewpoints.
- each viewpoint around the display sees a vertical line that scans out pixels from numerous projected MCOP images to form a single perspective image.
- the formation of these slices may be captured using a high-speed camera.
- the number of slices that make up an observed image depends on the viewpoint's distance from the display. This may be tested by projecting a sequence of alternating all-black and all-white images, allowing the number of images contributing to any one viewpoint to be counted easily. Closer to the mirror, the number of images that contributes to the view increases. As the viewpoint recedes, the number of images contributing to a view decreases to a minimum of approximately ten. This number never drops to one since our video projector is not orthographic.
- Simpler techniques than those described above may be used to project imagery to the display, but they typically do not achieve correct perspective.
- FIG. 4 illustrates a geometric calibration process for the system illustrated in FIG. 1 , in one embodiment of the present disclosure.
- the projection process described above requires knowledge of the intrinsic projector parameters and its pose relative to the spinning mirror.
- the world coordinates are chosen to originate at the center of the mirror, with the vertical axis (0,1,0) oriented along the mirror's axis of rotation. Calibration is relatively straightforward as only a single projector is used, and an optical path is used with a single rotating element.
- a simple and known linear calibration approach is used. This method requires at least 6 correspondences between known 3D points and their transformed 2D pixel positions. Radial lens distortion is ignored, in one embodiment, as this was measured to be insignificant.
- known 3D positions are indicated by marking fixed points on the mirror surface. With the motor off, the mirror is positioned so that it faces the front of the display. A paper calibration target is attached that consists of five fiducial markers ( 401 , 402 , 403 , 404 , 405 ) on a surface 407 of a mirror 410 , as seen FIG. 4 . The mirror 410 spins when actuated by a motor 415 , by analogy to the mirror 110 described in conjunction with FIG. 1 .
- a centered crosshair pattern is projected from the projector so that it can be positioned directly above the center fiducial.
- the video projector (not shown in FIG. 4 ; shown and labeled as 120 in FIG. 1 ) is mounted so that its central projected pixel projects down vertically.
- a mouse may be used to move the crosshair to each of the other fiducial markers, clicking the mouse to obtain the position of the corresponding projector pixel.
- the mirror 410 is then rotated 180°, and then the four fiducials are clicked again, obtaining a total of eight 2D points.
- the eight fiducial positions form a unit cube in space.
- 4 D four dimensional light fields may be captured, preprocessed, and dynamically rendered to the display system described above, with correct horizontal and vertical parallax.
- FIG. 5 is a schematic block diagram of a system 500 for dynamic rendering of a 3D display from a photographically captured 4D light field data set, in one embodiment of the present disclosure.
- the system 500 includes a video camera 510 , a motorized turntable 520 that rotates around an axis of rotation 522 when actuated, a high speed video projector 530 , a mirror 540 configured to spin when actuated by a motion-control motor 550 , and a processing system 560 that contains a graphics card 562 interfaced to the video projector 530 .
- the system 500 captures a 4D light field of a real object 580 (that is to be rendered).
- the object 580 is placed on the motorized turntable 520 .
- the video camera 510 is placed at a distance D in front of the object.
- D may be about 1.0 m, although D may of course be varied in different embodiments of the present disclosure.
- the object 580 is lit, for example with ambient light and/or lights attached to the turntable 520 or elsewhere, so that in one embodiment, the object 580 and its illumination remain substantially in the same relationship to each other during the rotation of the turntable 520 and the object 580 .
- the video camera 510 captures a movie sequence of at least 288 frames of the object 580 rotating 360° on the turntable 520 . This takes a few seconds.
- the system 500 may capture a full 4D light field by shooting (with the video camera 510 ) multiple rotations of the turntable 520 , raising the camera's height H by about 1.25 cm for each successive rotation.
- the intrinsic parameters for the camera may be calibrated, the and its pose may be recorded for each rotation.
- the processing system 560 pre-processes the light field to produce images appropriate for projection.
- the object 580 and display coordinate systems are aligned by placing the origin at a point within the center of the object directly above the center of the turntable.
- the y axis is aligned to the turntable's axis of rotation 522 .
- a new, rebinned, light field slice is generated as follows.
- the virtual viewing space or viewing circle V is placed around the display at height Hi and distance D.
- rays are traced from the reflected projector at P′ through each pixel (u, v) to the mirror at M through to the view-point V′ on V and then back toward M as described above in conjunction with FIGS. 3A-3C .
- the light field then simply needs to be queried for its radiance along ray.
- V′ thus lies on or between two of the same slice's camera locations Ci and Ci+1 as in FIG. 3C .
- To obtain the final pixel value one only needs to bilinearly interpolate between the pixels from Ci and Ci+1 that look toward point M on the mirror.
- the rebinned slices are then dithered to create binary images.
- Sets of 24 halftoned images are packed into 24-bit color images.
- 768 ⁇ 768 resolution one slice requires just over 20 MB of texture memory, allowing a light field resolution of over 768 ⁇ 768 pixels by 288 ⁇ 32 views to be stored on a modern 768 MB graphics card.
- each one of the rebinned light field slices yields correct perspective when projected on the display and observed anywhere from the original slice's height Hi and distance D. If the viewer distance remains near distance D, one could produce accurate vertical parallax by swapping which slice is displayed according to the user's height. To render the light field accurately for any height and distance, a dynamic rebinning process is used as described below.
- FIGS. 6A-6B illustrate dynamic vertical rebinning to generate correct vertical parallax for a horizontal-parallax-only 3D display of a photographically captured 4D light field data set, in one embodiment of the present disclosure.
- a dynamic vertical rebinning is performed that samples from different preprocessed light field slices based on the viewer's height h and distance d to produce correct horizontal and vertical perspective on the light field for any viewpoint.
- each slice i's nodal point is considered at distance D and height Hi in front of the mirror as shown in FIG. 6A .
- FIG. 6A shows vertical light field rebinning that is performed by projecting the light field slice closest in angle to the viewpoints ( 601 , 602 , 603 , 604 , 605 , 606 , and 607 ) onto each area of the mirror 610 .
- the midpoints between the slices are projected through the viewer position onto the mirror, and then up into the projector image.
- the regions are delimited as quadrilaterals ( 650 , 651 , and 652 ) that extend wide enough to cover the image as seen in FIG. 6B .
- FIG. 6B shows that these projected regions define textured quadrilaterals ( 650 , 651 , 652 ) on the mirror surface.
- a texture-mapped polygon is rendered that copies over the corresponding region from each light field slice.
- a projected image may be built up from these different slices.
- the areas corresponding to different original slices may be made visible by inverting every other quadrilateral of the dynamically rebinned projector frame.
- multiple temporal samples may be stored of a horizontal-parallax-only light field.
- photographs from a 25-frame animated light field may captured and rendered using a known flowed reflectance field rendering technique.
- light fields from multi-camera systems could be used, or a high-speed single-camera system using a spinning mirror to vary the viewpoint could be used to capture such data.
- a number of methods may be used to display color imagery.
- a straightforward method to create a color version of the above-described 3D display may use a 3-chip DMD projector.
- a two-channel field-sequential color system may be implemented using a two-sided tent-shaped diffusing mirror 710 shown in FIG. 7 .
- a color filter may be placed between a holographic diffusing film and a first-surface of the mirror 710 . This avoids introducing specular first-surface reflections.
- a Lee #131 cyan filter may be chosen for one side 720 of the mirror 710 and a Lee #020 orange filter (symbolically illustrated with stripes) for the other side 730 , dividing the visible spectrum approximately evenly into short and long wave-lengths.
- the RGB (red, blue and green) colors may be converted to Orange-Cyan colors by projecting the linear RGB vector onto the plane spanned by the Orange and Cyan colors.
- each plane of the tent mirror may be calibrated independently as described earlier. Then, the 3D scene may be rendered twice for each sub-frame, once for the orange side and once for the cyan side.
- the calibration process ensures that each side is rendered toward the appropriate set of viewpoints.
- the effect for the viewer is similar to the Kinemacolor 2-color cinema system, and the choice of filters allows for useful color reproduction for many scenes.
- the tent-mirror system doubles the number of images per second shown to the viewers, allowing a 40 Hz field-sequential color frame rate which appears significantly more stable than 20 Hz monochrome. Additional color channels and/or update rates may be achieved using additional mirrors in the arrangement, making the set of mirrors resemble a multi-sided pyramid.
- the projection techniques described in the present disclosure allow correct vertical perspective and parallax to be rendered for any height and distance when these parameters are known, and this effect may be demonstrated with interactive raster graphics using a tracking system to measure the viewer's height and distance. Finally, the projection techniques described in the present disclosure have been applied to the display of photographed light fields with accurate horizontal and vertical parallax.
Abstract
An interactive, autostereoscopic system for displaying an object in 3D includes a mirror configured to spin around a vertical axis when actuated by a motor, a high speed video projector, and a processing system including a graphics card interfaced to the video projector. An anisotropic reflector is bonded onto an inclined surface of the mirror. The video projector projects video signals of the object from the projector onto the inclined surface of the mirror while the mirror is spinning, so that light rays representing the video signals are redirected toward a field of view of a 360 degree range. The processing system renders the redirected light rays so as to interactively generate a horizontal-parallax 3D display of the object. Vertical parallax can be included in the display by adjusting vertically the displayed views of the object, in response to tracking of viewer motion by a tracking system.
Description
- This application is a continuation application of U.S. patent application Ser. No. 12/105,120, entitled “Rendering for an Interactive 360 Degree Light Field Display,” filed Apr. 17, 2008, attorney docket no. 028080-0334; which is based upon and claims priority to U.S. Provisional Application No. 60/912,370 entitled “An Interactive 360 Degree Horizontal Parallax Light Field Display,” filed Apr. 17, 2007, attorney docket no. 028080-0270. The entire content of both of these applications is incorporated herein by reference.
- This invention was made with government support under Grant No. W911 NF-04-D-0005, awarded by the U.S. Army Research Office. The government has certain rights in the invention.
- A great deal of computer generated imagery is modeled and rendered in 3D (three dimensions). The vast majority of 3D imagery is shown on 2D displays, however. Only relatively recently have the requisite enabling technologies become available that can offer the promise of making functional and practical 3D displays possible.
- Many problems and challenges remain in this field. As one example, rendering 3D displays of scenes with both the correct horizontal and vertical perspectives remains a challenge. Another problem that has not yet been addressed is how to achieve 3D displays that are interactive so as to allow real time updates to the displayed images.
- There is a need for improved methods and systems for 3D displays that can among other things address the problems and challenges mentioned above.
- An interactive, autostereoscopic system for displaying an object in three dimensions includes a mirror configured to spin around a vertical axis when actuated by a motor, a high speed video projector, and a processing system including a graphics card interfaced to the video projector. An anisotropic reflector is bonded onto an inclined surface of the mirror. The video projector projects video signals of the object from the projector onto the projection surface while the mirror is spinning, so that light rays representing the video signals are redirected toward a field of view of a 360 degree range within a viewing space of one or more viewers. The processing system renders the redirected light rays so as to generate a horizontal-
parallax 3D display of the object for the viewers from any of a plurality of viewing directions within the viewing space. -
FIG. 1 illustrates a system for generating an autostereoscopic, multiview (horizontal-parallax-only) 3D display of an object, in one embodiment of the present disclosure. -
FIG. 2 illustrates the anisotropic reflectance characteristics of the mirror system illustrated inFIG. 1 . -
FIGS. 3A , 3B, and 3C illustrate a projection process and geometry for rendering an object or scene to the system illustrated inFIG. 1 , in one embodiment of the present disclosure. -
FIG. 4 illustrates a geometric calibration for the system illustrated inFIG. 1 , in one embodiment of the present disclosure. -
FIG. 5 illustrates a system for dynamic rendering of a 3D display from a photographically captured 4D light field data set, in one embodiment of the present disclosure. -
FIGS. 6A-6B illustrate dynamic vertical rebinning to generate correct vertical parallax for a horizontal-parallax-only 3D display of a photographically captured 4D light field data set, in one embodiment of the present disclosure. -
FIG. 7 illustrates a two-mirror tent for displaying two-toned color imagery using orange and cyan filters below the diffusers. - The present disclosure describes an easily reproducible, low-
cost 3D display system with a form factor for displaying 3D objects in 3D. Also described are the projection mathematics and rendering methods necessary to drive the 3D display with real time raster imagery or pre-recorded light fields so that they exhibit the correct cues of both horizontal and vertical parallax. The display is autostereoscopic, requiring no special viewing glasses, and omnidirectional, allowing viewers to be situated anywhere around it. The display is also multiview, i.e. produces a correct rendition of the light field with the correct horizontal parallax and vertical perspective for any viewpoint situated at a certain distance and height around the display. If head tracking is employed to detect the height and distance of one or more of the viewers around the display, the projection equations described below allow the vertical parallax to be adjusted on the fly to allow the tracked users to properly see objects from arbitrary heights and distances, in addition to obtaining correct views from any angle around the display. The systems and methods described below use primarily commodity graphics and display components and achieve real-time rendering with non-trivial scene complexity across its entire field of view. -
FIG. 1 illustrates a system for generating an autostereoscopic 3D display of an object, in one embodiment of the present disclosure. The3D display system 100 includes: amirror 110 covered by ananisotropic reflector 112, a motion-control motor 115, a high-speed video projector 120, and a processing system 130 (for example a standard PC) that includes agraphics card 132 interfaced to thevideo projector 120 using a custom FPGA-based image decoder. Themirror 110 is configured to spin with respect to a vertical axis, when actuated by the motion-control motor 115. Thespinning mirror 110 may be tilted or inclined at an angle with respect to the vertical axis, for example by about 45 degrees, to reflect rays of light from theprojector 120 to all possible viewing positions around the device, allowing many viewers to view the display simultaneously. - In one embodiment, illustrated in
FIG. 1 , thevideo projector 120 achieves high-speed projection by modifying an off-the-shelf projector to use a new DLP drive card with custom programmed FPGA-based circuitry. The FPGA decodes a standard DVI signal from thegraphics card 132. Instead of rendering a color image, the FPGA takes each 24-bit color frame of video and displays each bit sequentially as separate frames. Thus, if the incoming digital video signal is 60 Hz, thevideo projector 120 displays 60×24=1,440 frames per second. To achieve even faster rates, the video card refresh may be set to rates of about 180-240 Hz. At 200 Hz, thevideo projector 120 displays 4,800 binary frames per second. The processing system 130 (containing the graphics card 132) continuously renders new horizontal views of the subject, which in the illustrated embodiment consists of 288 images per rotation of the mirror. These views are encoded into 24-bit images and sent to thevideo projector 120. A complete kit consisting of the FPGA and DLP boards is now available from Polaris Road, Inc. - The
anisotropic reflector 112 is a holographic diffuser is bonded onto a surface of themirror 110. In one embodiment, illustrated inFIG. 1 , theanisotropic reflector 112 is a holographic diffuser. Other embodiments of the present disclosure may use different types of anisotropic reflectors, including but not limited to: a lenticular lens; and a lenslet array. The mirror surface (with the anisotropic reflector bonded thereon) provides aprojection surface 114, which reflects each projector pixel to a narrow range of viewpoints. In other words, the mirror surface (onto which the anisotropic reflector bonded) provides a projection surface that reflects incident rays of light in a way that the rays are spread in a substantially vertical direction after being reflected. This configuration contrasts with previous volumetric displays, which projected images onto a spinning diffuse plane which scattered light in all directions. Such displays could not recreate view-dependent effects such as occlusion. - The
holographic diffuser 112 provides control over the width and height of this region. In the illustrated embodiment, the characteristics of theholographic diffuser 112 are such that the relative diffusion between x and y is approximately 1:200. Horizontally, thesurface 114 is sharply specular to maintain a 1.25 degree separation between views. Vertically, themirror 110 scatters widely so the projected image can be viewed from essentially any height. -
FIG. 2 shows the anisotropic reflectance characteristics of the mirror system (i.e. themirror 110 coupled to themotor 115 and having theholographic diffuser 112 bonded to a surface of the mirror 110). The horizontal profile of the specular lobe approximates a bilinear interpolation between adjacent viewpoints. The motion of the mirror may add some additional blur which improves reproduction of halftoned imagery at the expense of angular resolution. The holographic diffuser may be diffuse in the vertical dimension and sharply specular in the horizontal dimension. In the graph shown inFIG. 2 , the dotted line plots the intensity of a laser beam as reflected by theholographic diffuser 112 andmirror 110 toward the viewers. The solid line illustrates the intensity of a thin vertical line of light from the video projector as reflected by theholographic diffuser 112 andmirror 110 toward the viewers. Finally, the dashed line illustrates the intensity of an ideal bilinear interpolation spread of a hat function whose radius matches the 1.25 degree angular separation of the successive views of thedisplay system 100. - The anisotropic
holographic diffuser 112 andmirror 110 may be mounted on a carbon fiber panel and attached to an aluminum fly-wheel at 45°, in the illustrated embodiment. The flywheel spins synchronously relative to the images displayed by thevideo projector 120. - In one embodiment, the
system 100 may be synchronized as follows. Since the output frame rate of the PC graphics card is relatively constant and cannot be fine-tuned on the fly, we use the PC video output rate as the master signal for system synchronization. The projector's FPGA also creates signals encoding the current frame rate. These control signals inter-face directly to an Animatics SM3420D “Smart Motor” which contains firmware and motion control parameters resulting in a stable, velocity-based control loop that ensures the motor velocity stays in sync with the signals from the projector. As the mirror rotates up to 20 times per second, persistence of vision creates the illusion of a floating object around the center of the mirror. - The projector and spinning mirror yield a horizontal-parallax-only display; the image perspective does not change correctly as the viewpoint moves up and down, or forward and backward. However, the projection algorithms described further below can take into account the height and distance of the viewer to render the scene with correct perspective.
- If just horizontal parallax is required, a sensible course of action is to initialize this height and distance to the expected typical viewing height and distance. Since the display system described above can be implemented interactively, using the graphics card we can achieve both horizontal and vertical parallax display by using a tracking system to measure the user's height and distance.
- The 3D, horizontal-parallax-only display may be augmented to include correct vertical parallax in the display. In overview, correct vertical parallax may be included by tracking the motion of one or more viewers, then interactively updating the displayed images of the object in response to the tracked motion of the one or more viewers. In an embodiment in which the motion of one viewer is tracked, the displayed images of the object may be interactively updated in such a way that when the viewer is moving upward from below, the displayed image of the object appears to the viewer as if seen from above, and when the viewer is moving downward from above the displayed image of the object appears to the viewer as if seen from below. In an embodiment in which the motions of a plurality of viewers are tracked, the displayed images of the object may be interactively updated in such a way that the images of the object as displayed to each viewer is shown from a viewpoint consistent with the height of each viewer relative to the display.
- To include correct vertical parallax, a tracking system is thus used, for example where the viewer holds the tracking device (a sensor) to their temple, or to a video camera filming the display. Any motion tracking system known in the art may be used. In one embodiment, the tracking system may be an electromagnetic motion tracking device, such as a Polhemus Patriot electromagnetic tracking device that can track the motion of a viewer when held by or attached to the viewer. In another embodiment, the tracking system may include one or more video cameras. In this embodiment, a plurality of video cameras may be used to observe a plurality of viewers, so that the position of each of the viewers can be determined. In yet another embodiment, the tracking system may include a plurality of depth cameras used to track the position of a plurality of viewers.
- The tracking data may used by the projection algorithm (further described below) to display the scene from the correct perspective for the viewer's height and distance. In this way, the display's horizontal parallax provides binocular stereo and yields zero lag as the viewer moves his/her head horizontally, which may be one of the most common significant head motions. The effects of vertical motion and distance change are computed based on the tracked position. The display only needs to adjust the rendered views in the vicinity of each tracked user, leaving the rest of the displayed circumference optimized to the average expected viewer position.
-
FIGS. 3A , 3B, and 3C illustrate projection process and geometry for rendering an object and/or a scene to the system illustrated inFIG. 1 , in one embodiment of the present disclosure. - In these figures, it is assumed that the spinning mirror (labeled as 110 in
FIG. 1 , re-labeled as 310 inFIG. 3A ) is centered at the origin and that its axis of rotation is the vertical y-axis, with the video projector at the nodal point P above the mirror, as shown inFIG. 3A . It is further assumed that the viewpoint for which the correct perspective should be obtained is at a height h (relative to the center of the projection surface 114) and a distance d from the y- or vertical axis about which themirror 310 rotates when actuated by the motion-control motor. - By the rotational symmetry of the 3D system described above, perspective-correct imagery can be produced for any viewing position on a viewing space or viewing circle V defined by h and d. This yields binocular images for a viewer facing the display since h and d will be similar for both eyes. A particular viewpoint on the circle V is denoted in the present disclosure as V′. In practice, the set of perspective-correct viewpoints V need not be a continuous planar circle and can pass through a variety of tracked viewer positions at different distances and heights.
- At any given instant, with the spinning
anisotropic mirror 310 frozen at a particular position, the 2D image projected onto the mirror is reflected out into space, covering parts of the field of view of many viewpoints on the viewing space V, as shown inFIG. 3B . Since themirror 310 provides little horizontal diffusion, each projector pixel (u, v) sends light toward one specific viewpoint V′ on V. The display system must ensure that each projected pixel displays the appropriate part of the scene as it should be seen from viewpoint V′. Thus, there are two questions that must be answered: First, for a 3D point Q in a scene, what is the corresponding projector pixel (u, v) that reflects to the correct viewpoint V′ along the ray ? Second, for a given projector pixel (u, v), which ray should be traced into the scene so that the display projects the correct ray intensity? The first answer tells us how to render 3D geometric models to the display and the second answer tells us how to render ray-traceable scenes such as light fields. - Projecting from the Scene into the Projector
- If the scene being rendered is a polygonal 3D model, it must be determined, for any world-space vertex Q, where it should be rendered on the projector's image for any given mirror position. To visualize this process, the 3D display system is considered as it would be viewed from above. It is noted that in the horizontal plane, the anisotropic mirror behaves in essence like a regular mirror. The optical path can thus be unfolded by reflecting the projector position P to P′ across the plane of the
mirror 310, as seen inFIG. 3B . A ray originating at P′ passing through Q will continue out into space toward the viewers. This ray will not, in general, intersect the view circle V. By assuming that the mirror diffuses rays into a vertical plane, the vertical plane containing is intersected with the viewing circle V, to determine the viewpoint V′ from which Q will be seen with the mirror at its given position. As explained in Appendix A to this disclosure, this diffusion plane is an approximation to a cone-shaped reflection from the mirror, but the projection error is small for the setup described above, and may be neglected in practice. - A ray is then traced from the viewpoint V′ toward Q, until it intersects the surface of the mirror at M. The symbol M represents the one point on the mirror that reflects light to the viewer coming from the direction of Q. To draw onto this point from the projector, M simply needs to be projected up toward the projector's nodal point P to find the corresponding projector pixel (u, v). Thus, illuminating a pixel at (u,v) will make it appear from viewpoint V′ that 3D point Q has been illuminated. Q will eventually be rendered as it should be seen from all other viewpoints on V as the mirror rotates.
- With these few geometric intersections, it can be determined for any 3D point Q where it should be drawn on the projector for each position of the mirror. Seen on the display by a viewer, the observed images exhibit correct perspective projection. The above-described technique actually renders multiple-center-of-projection (MCOP) images to the projector which can not be generated using a traditional projection matrix. Specifically, the projection technique implemented by the processing system 130 (shown in
FIG. 1 ) uses a combination of two different viewpoints P (for horizontal coordinates) and V′ (for vertical coordinates). Nonetheless, the technique may easily be implemented as the vertex shader provided in Appendix B of the present disclosure, allowing an entire mesh to be rendered in a single pass. For z-buffering, vertex depth may be based on the distance from V′ to Q. - In the above-described MCOP projection, long straight lines should naturally appear curved in the projection. Thus, models with large polygons should be tesselated; alternatively, a fragment shader could discard incorrect pixels that lie outside the triangle.
- Ray Tracing from the Projector into the Scene
- If the scene to be displayed (such as a light field) is most easily ray-traced, it must be determined for each projector pixel (u, v) which ray in space—from the viewer toward the scene—corresponds to that pixel. The reflected projector position in
FIG. 3B is again used, and a ray is projected from P′ through its corresponding pixel (u, v) to where it intersects the surface of the mirror at point M. Upon intersecting the diffuser, it is assumed that this ray P′M spreads into a vertical fan of light which intersects the circle of views V at V′. Seen from above, this intersection is easily calculated as a 2D line-circle intersection. - It can now be seen that projector pixel (u, v) reflects from mirror point M toward viewpoint V′. Thus, the color it should display should be the result of tracing a ray from V′ toward point M. If the scene being rendered is a light field, ray V′M is simply queried for the scene radiance at that point, as further described below.
- The fans of light from a given projector frame diverge horizontally toward multiple viewpoints. As the mirror rotates, each viewpoint around the display sees a vertical line that scans out pixels from numerous projected MCOP images to form a single perspective image. The formation of these slices may be captured using a high-speed camera. The number of slices that make up an observed image depends on the viewpoint's distance from the display. This may be tested by projecting a sequence of alternating all-black and all-white images, allowing the number of images contributing to any one viewpoint to be counted easily. Closer to the mirror, the number of images that contributes to the view increases. As the viewpoint recedes, the number of images contributing to a view decreases to a minimum of approximately ten. This number never drops to one since our video projector is not orthographic.
- Simpler techniques than those described above may be used to project imagery to the display, but they typically do not achieve correct perspective.
-
FIG. 4 illustrates a geometric calibration process for the system illustrated inFIG. 1 , in one embodiment of the present disclosure. - The projection process described above (in conjunction with
FIGS. 3A , 3B, and 3C) requires knowledge of the intrinsic projector parameters and its pose relative to the spinning mirror. The world coordinates are chosen to originate at the center of the mirror, with the vertical axis (0,1,0) oriented along the mirror's axis of rotation. Calibration is relatively straightforward as only a single projector is used, and an optical path is used with a single rotating element. - In the embodiment illustrated in
FIG. 4 , a simple and known linear calibration approach is used. This method requires at least 6 correspondences between known 3D points and their transformed 2D pixel positions. Radial lens distortion is ignored, in one embodiment, as this was measured to be insignificant. In this approach, known 3D positions are indicated by marking fixed points on the mirror surface. With the motor off, the mirror is positioned so that it faces the front of the display. A paper calibration target is attached that consists of five fiducial markers (401, 402, 403, 404, 405) on asurface 407 of amirror 410, as seenFIG. 4 . Themirror 410 spins when actuated by amotor 415, by analogy to themirror 110 described in conjunction withFIG. 1 . - A centered crosshair pattern is projected from the projector so that it can be positioned directly above the center fiducial. In this embodiment, the video projector (not shown in
FIG. 4 ; shown and labeled as 120 inFIG. 1 ) is mounted so that its central projected pixel projects down vertically. A mouse may be used to move the crosshair to each of the other fiducial markers, clicking the mouse to obtain the position of the corresponding projector pixel. Themirror 410 is then rotated 180°, and then the four fiducials are clicked again, obtaining a total of eight 2D points. The eight fiducial positions form a unit cube in space. - Dynamic Rendering of 3D Display from 4D Light Field Data Set
- Using the ray tracing projection described above in conjunction with
FIGS. 3A , 3B, and 3C, 4D (four dimensional) light fields may be captured, preprocessed, and dynamically rendered to the display system described above, with correct horizontal and vertical parallax. -
FIG. 5 is a schematic block diagram of asystem 500 for dynamic rendering of a 3D display from a photographically captured 4D light field data set, in one embodiment of the present disclosure. In overview, thesystem 500 includes avideo camera 510, amotorized turntable 520 that rotates around an axis of rotation 522 when actuated, a highspeed video projector 530, amirror 540 configured to spin when actuated by a motion-control motor 550, and aprocessing system 560 that contains agraphics card 562 interfaced to thevideo projector 530. - As a first step, the
system 500 captures a 4D light field of a real object 580 (that is to be rendered). In the illustrated embodiment, theobject 580 is placed on themotorized turntable 520. Thevideo camera 510 is placed at a distance D in front of the object. In the illustrated embodiment, D may be about 1.0 m, although D may of course be varied in different embodiments of the present disclosure. Theobject 580 is lit, for example with ambient light and/or lights attached to theturntable 520 or elsewhere, so that in one embodiment, theobject 580 and its illumination remain substantially in the same relationship to each other during the rotation of theturntable 520 and theobject 580. - In the illustrated embodiment, the
video camera 510 captures a movie sequence of at least 288 frames of theobject 580 rotating 360° on theturntable 520. This takes a few seconds. In one embodiment, thesystem 500 may capture a full 4D light field by shooting (with the video camera 510) multiple rotations of theturntable 520, raising the camera's height H by about 1.25 cm for each successive rotation. The intrinsic parameters for the camera may be calibrated, the and its pose may be recorded for each rotation. - As explained above, regular perspective images shown directly on the projector may not produce correct perspective to viewers around the display. Thus, the
processing system 560 pre-processes the light field to produce images appropriate for projection. First, theobject 580 and display coordinate systems are aligned by placing the origin at a point within the center of the object directly above the center of the turntable. The y axis is aligned to the turntable's axis of rotation 522. - Then, for each slice i of the captured light field taken from height Hi, a new, rebinned, light field slice is generated as follows. The virtual viewing space or viewing circle V is placed around the display at height Hi and distance D. Then, for each of the 288 mirror positions, rays are traced from the reflected projector at P′ through each pixel (u, v) to the mirror at M through to the view-point V′ on V and then back toward M as described above in conjunction with
FIGS. 3A-3C . The light field then simply needs to be queried for its radiance along ray. This is a simple query since V was chosen to be coincident with the height and distance of the current slice of the light field: V′ thus lies on or between two of the same slice's camera locations Ci and Ci+1 as inFIG. 3C . To obtain the final pixel value, one only needs to bilinearly interpolate between the pixels from Ci and Ci+1 that look toward point M on the mirror. - For the display, the rebinned slices are then dithered to create binary images. Sets of 24 halftoned images are packed into 24-bit color images. As there are 288 images in each rebinned slice, this yields twelve 24-bit color images per row. At 768×768 resolution, one slice requires just over 20 MB of texture memory, allowing a light field resolution of over 768×768 pixels by 288×32 views to be stored on a modern 768 MB graphics card.
- By construction, each one of the rebinned light field slices yields correct perspective when projected on the display and observed anywhere from the original slice's height Hi and distance D. If the viewer distance remains near distance D, one could produce accurate vertical parallax by swapping which slice is displayed according to the user's height. To render the light field accurately for any height and distance, a dynamic rebinning process is used as described below.
-
FIGS. 6A-6B illustrate dynamic vertical rebinning to generate correct vertical parallax for a horizontal-parallax-only 3D display of a photographically captured 4D light field data set, in one embodiment of the present disclosure. - In the illustrated embodiment, a dynamic vertical rebinning is performed that samples from different preprocessed light field slices based on the viewer's height h and distance d to produce correct horizontal and vertical perspective on the light field for any viewpoint. For each mirror position, each slice i's nodal point is considered at distance D and height Hi in front of the mirror as shown in
FIG. 6A .FIG. 6A shows vertical light field rebinning that is performed by projecting the light field slice closest in angle to the viewpoints (601, 602, 603, 604, 605, 606, and 607) onto each area of themirror 610. The midpoints between the slices are projected through the viewer position onto the mirror, and then up into the projector image. These projected midpoints form an axis of points crossing the center of the projector image. Lines from each point are extended perpendicularly to this axis, dividing the projector's image into a set of regions orareas 650, each one corresponding to the area for which light field slice i contains the rays that most closely correspond to the viewpoint's view of the scene over that area. - The regions are delimited as quadrilaterals (650, 651, and 652) that extend wide enough to cover the image as seen in
FIG. 6B .FIG. 6B shows that these projected regions define textured quadrilaterals (650, 651, 652) on the mirror surface. For each quadrilateral, a texture-mapped polygon is rendered that copies over the corresponding region from each light field slice. A projected image may be built up from these different slices. The areas corresponding to different original slices may be made visible by inverting every other quadrilateral of the dynamically rebinned projector frame. - If the viewer is close to distance D from the display, just one or two light field slices will constitute the projected images. As the viewer moves forward or back from D, the number of slices used will increase. Since the images on the graphics card are already dithered, no blending is performed between the slices. However, the seams between the slices may not be noticeable if the light field is of sufficient vertical angular resolution.
- Instead of using the graphics card's memory to store multiple vertical slices of an object's light field, multiple temporal samples may be stored of a horizontal-parallax-only light field. For example, photographs from a 25-frame animated light field may captured and rendered using a known flowed reflectance field rendering technique. Alternatively, light fields from multi-camera systems could be used, or a high-speed single-camera system using a spinning mirror to vary the viewpoint could be used to capture such data.
- A number of methods may be used to display color imagery. A straightforward method to create a color version of the above-described 3D display may use a 3-chip DMD projector.
- Alternatively, a two-channel field-sequential color system may be implemented using a two-sided tent-shaped
diffusing mirror 710 shown inFIG. 7 . For each side of the tent, a color filter may be placed between a holographic diffusing film and a first-surface of themirror 710. This avoids introducing specular first-surface reflections. In one exemplary embodiment, a Lee #131 cyan filter may be chosen for oneside 720 of themirror 710 and a Lee #020 orange filter (symbolically illustrated with stripes) for theother side 730, dividing the visible spectrum approximately evenly into short and long wave-lengths. The RGB (red, blue and green) colors may be converted to Orange-Cyan colors by projecting the linear RGB vector onto the plane spanned by the Orange and Cyan colors. - To render in color, each plane of the tent mirror may be calibrated independently as described earlier. Then, the 3D scene may be rendered twice for each sub-frame, once for the orange side and once for the cyan side. The calibration process ensures that each side is rendered toward the appropriate set of viewpoints. The effect for the viewer is similar to the Kinemacolor 2-color cinema system, and the choice of filters allows for useful color reproduction for many scenes. Besides achieving color, the tent-mirror system doubles the number of images per second shown to the viewers, allowing a 40 Hz field-sequential color frame rate which appears significantly more stable than 20 Hz monochrome. Additional color channels and/or update rates may be achieved using additional mirrors in the arrangement, making the set of mirrors resemble a multi-sided pyramid.
- In sum, methods and systems have been described that allow an autostereoscopic light field display to present interactive 3D graphics to multiple simultaneous viewers 360 degrees around the display, including rendering techniques used to generate the imagery on the display. Using a standard programmable graphics card, over 5,000 images per second of interactive 3D graphics may be rendered, projecting 360-degree views with 1.25 degree separation up to 20 updates per second. The system's projection geometry and its calibration process have been described. A multiple-center-of-projection rendering technique has been presented for creating perspective-correct images from arbitrary viewpoints around the display.
- The projection techniques described in the present disclosure allow correct vertical perspective and parallax to be rendered for any height and distance when these parameters are known, and this effect may be demonstrated with interactive raster graphics using a tracking system to measure the viewer's height and distance. Finally, the projection techniques described in the present disclosure have been applied to the display of photographed light fields with accurate horizontal and vertical parallax.
- The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of what is disclosed above. Thus, the apparatuses and methods described above are not intended to be limited to the embodiments shown herein, but are to be accorded the full scope consistent with the claims, wherein reference to an element in the singular is not intended to mean “one and only one” unless specifically so stated, but rather “one or more.”
- All structural and functional equivalents to the elements of the various embodiments described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference, and are intended to be encompassed by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether or not such disclosure is explicitly recited in the claims. No claim element is to be construed under the provisions of 35 U.S.C. §112, sixth paragraph, unless the element is expressly recited using the phrase “means for” or, in the case of a method claim, the element is recited using the phrase “step for.”
Claims (11)
1. (canceled)
2. A system for generating a 3D display of an object in 3D space, comprising:
a display system that renders multiple different images of the object, each different image being projected from the display system at substantially the same time at a different angle within a first viewing plane and representing how the object would look if actually viewed from that angle within the first viewing plane, without the viewer having to wear any special glasses or headgear;
a tracking system that tracks the position of a viewer of one of the images in a second viewing plane that is substantially perpendicular to the first viewing plane; and
a processing system that receives information from the tracking system indicative of the position of the viewer in the second viewing plane and, in response, causes the display system to render the image of the object that the viewer is viewing as the object would look if actually viewed from the position of the viewer within the second viewing plane, without the viewer having to wear any special glasses or headgear.
3. The system of claim 2 wherein the first viewing plane is substantially horizontal and the second viewing plane is substantially vertical.
4. The system of claim 2 wherein each image of the object is an autostereoscopic image of the object.
5. The system of claim 2 wherein:
the tracking system is configured to track the positions of multiple viewers in the second viewing plane substantially simultaneously, each while at different positions in the second viewing plane; and
the processing system receives information from the tracking system indicative of the position of each of the viewers in the second viewing plane and, in response, causes the display system to render the image of the object that each viewer is viewing as the object would look if actually viewed from the position of the viewer within the second viewing plane, without any of the viewers having to wear any special glasses or headgear.
6. The system of claim 5 wherein the first viewing plane is substantially horizontal and the second viewing plane is substantially vertical.
7. The system of claim 5 wherein the tracking system includes one or more video cameras configured to observe the multiple viewers.
8. The system of claim 5 wherein the tracking system includes one or more depth cameras configured to track the position of the multiple viewers.
9. The system of claim 2 wherein the different angles within the first viewing plane span a range of approximately 360 degrees.
10. The system of claim 2 wherein the display system includes a rotating anisotropic surface.
11. The system of claim 2 wherein the tracking system includes a magnetic tracking device configured to track the position of the viewer.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/850,229 US20140327748A1 (en) | 2007-04-17 | 2013-03-25 | Rendering for an interactive 360 degree light field display |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US91237007P | 2007-04-17 | 2007-04-17 | |
US12/105,120 US8432436B2 (en) | 2007-04-17 | 2008-04-17 | Rendering for an interactive 360 degree light field display |
US13/850,229 US20140327748A1 (en) | 2007-04-17 | 2013-03-25 | Rendering for an interactive 360 degree light field display |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/105,120 Continuation US8432436B2 (en) | 2007-04-17 | 2008-04-17 | Rendering for an interactive 360 degree light field display |
Publications (1)
Publication Number | Publication Date |
---|---|
US20140327748A1 true US20140327748A1 (en) | 2014-11-06 |
Family
ID=40087662
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/105,120 Active 2031-02-06 US8432436B2 (en) | 2007-04-17 | 2008-04-17 | Rendering for an interactive 360 degree light field display |
US13/850,229 Abandoned US20140327748A1 (en) | 2007-04-17 | 2013-03-25 | Rendering for an interactive 360 degree light field display |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/105,120 Active 2031-02-06 US8432436B2 (en) | 2007-04-17 | 2008-04-17 | Rendering for an interactive 360 degree light field display |
Country Status (2)
Country | Link |
---|---|
US (2) | US8432436B2 (en) |
WO (1) | WO2009011946A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9927628B2 (en) | 2014-07-18 | 2018-03-27 | National Institute Of Information And Communications Technology | Image display apparatus |
US10134190B2 (en) | 2016-06-14 | 2018-11-20 | Microsoft Technology Licensing, Llc | User-height-based rendering system for augmented reality objects |
US10585393B2 (en) | 2015-08-28 | 2020-03-10 | Electronics And Telecommunications Research Institute | Apparatus for holographic display |
Families Citing this family (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9123116B2 (en) | 2011-04-19 | 2015-09-01 | University Of Southern California | Multiview face capture using polarized spherical gradient illumination |
US8330802B2 (en) * | 2008-12-09 | 2012-12-11 | Microsoft Corp. | Stereo movie editing |
US9648346B2 (en) * | 2009-06-25 | 2017-05-09 | Microsoft Technology Licensing, Llc | Multi-view video compression and streaming based on viewpoints of remote viewer |
US9733699B2 (en) * | 2010-04-13 | 2017-08-15 | Dean Stark | Virtual anamorphic product display with viewer height detection |
US8908013B2 (en) * | 2011-01-20 | 2014-12-09 | Canon Kabushiki Kaisha | Systems and methods for collaborative image capturing |
US8884949B1 (en) | 2011-06-06 | 2014-11-11 | Thibault Lambert | Method and system for real time rendering of objects from a low resolution depth camera |
US8922591B2 (en) | 2011-08-25 | 2014-12-30 | Innovare Solutions, LLC | Holographic display |
US9182524B2 (en) * | 2011-09-29 | 2015-11-10 | Disney Enterprises, Inc. | Autostereoscopic display system with one dimensional (1D) retroreflective screen |
US8746889B2 (en) | 2011-10-27 | 2014-06-10 | Delphi Technologies, Inc. | Auto-variable perspective autostereoscopic 3D display |
EP2915324B1 (en) | 2012-11-05 | 2020-07-08 | 360 Heros, Inc. | 360 degree camera mount and related photographic and video system |
JP6143469B2 (en) * | 2013-01-17 | 2017-06-07 | キヤノン株式会社 | Information processing apparatus, information processing method, and program |
KR102049456B1 (en) * | 2013-04-05 | 2019-11-27 | 삼성전자주식회사 | Method and apparatus for formating light field image |
US9367221B2 (en) * | 2013-09-02 | 2016-06-14 | Ken Lai | System and method for sequencing rotatable images |
US10116914B2 (en) * | 2013-10-31 | 2018-10-30 | 3Di Llc | Stereoscopic display |
US9883173B2 (en) | 2013-12-25 | 2018-01-30 | 3Di Llc | Stereoscopic display |
US9986228B2 (en) | 2016-03-24 | 2018-05-29 | 3Di Llc | Trackable glasses system that provides multiple views of a shared display |
US10652525B2 (en) | 2013-10-31 | 2020-05-12 | 3Di Llc | Quad view display system |
US11343487B2 (en) | 2013-10-31 | 2022-05-24 | David Woods | Trackable glasses system for perspective views of a display |
JP2015132953A (en) * | 2014-01-10 | 2015-07-23 | キヤノン株式会社 | Image processor and method thereof |
CN103744257B (en) * | 2014-01-10 | 2016-03-09 | 苏州博众机器人有限公司 | A kind of mechanism coordinating camera to realize frame freely to locate |
EP3099076B1 (en) * | 2015-05-29 | 2019-08-07 | InterDigital CE Patent Holdings | Method for displaying a content from 4d light field data |
US9549174B1 (en) | 2015-10-14 | 2017-01-17 | Zspace, Inc. | Head tracked stereoscopic display system that uses light field type data |
CN105467809A (en) * | 2015-12-15 | 2016-04-06 | 重庆触视科技有限公司 | Distortion-proof 3D holographic projection apparatus |
GB2546505B (en) * | 2016-01-19 | 2020-02-19 | Bruce Helson Christopher | Image display screen comprising rotating screen with optically anisotropic reflective layer |
JP7278277B2 (en) | 2017-11-02 | 2023-05-19 | ピーシーエムエス ホールディングス インコーポレイテッド | Method and system for aperture enlargement in light field displays |
RU2686576C1 (en) | 2017-11-30 | 2019-04-29 | Самсунг Электроникс Ко., Лтд. | Holographic display compact device |
US10869022B2 (en) * | 2018-03-21 | 2020-12-15 | City University Of Hong Kong | Autostereoscopic multi-view display system and related apparatus |
CN110088820B (en) * | 2019-03-11 | 2021-12-03 | 京东方科技集团股份有限公司 | Display device and operation method thereof |
MX2022000041A (en) | 2019-06-28 | 2022-04-06 | Pcms Holdings Inc | Optical method and system for light field (lf) displays based on tunable liquid crystal (lc) diffusers. |
EP4010756A4 (en) * | 2019-08-09 | 2023-09-20 | Light Field Lab, Inc. | Light field display system based digital signage system |
US11163178B1 (en) | 2020-04-17 | 2021-11-02 | Toyota Motor Engineering And Manufacturing North America, Inc. | Volumetric display using noble gasses |
CN115691383B (en) * | 2022-09-27 | 2023-12-22 | 惠科股份有限公司 | Display device and image display method thereof |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5365370A (en) * | 1993-06-10 | 1994-11-15 | Hudgins J Stephen | Three dimensional viewing illusion with 2D display |
US6744569B2 (en) * | 2001-06-19 | 2004-06-01 | Genex Technologies, Inc | Method and apparatus for omnidirectional three dimensional imaging |
US20070121182A1 (en) * | 2005-09-29 | 2007-05-31 | Rieko Fukushima | Multi-viewpoint image generation apparatus, multi-viewpoint image generation method, and multi-viewpoint image generation program |
US20070146358A1 (en) * | 2004-01-09 | 2007-06-28 | Koninklijke Philips Electronic, N.V. | Three-dimensional display |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5870220A (en) * | 1996-07-12 | 1999-02-09 | Real-Time Geometry Corporation | Portable 3-D scanning system and method for rapid shape digitizing and adaptive mesh generation |
GB9713658D0 (en) * | 1997-06-28 | 1997-09-03 | Travis Adrian R L | View-sequential holographic display |
US6648475B1 (en) * | 2002-05-20 | 2003-11-18 | Eastman Kodak Company | Method and apparatus for increasing color gamut of a display |
WO2004072659A2 (en) | 2003-02-07 | 2004-08-26 | University Of Southern California | Reflectometry apparatus and method |
US7436403B2 (en) | 2004-06-12 | 2008-10-14 | University Of Southern California | Performance relighting and reflectance transformation with time-multiplexed illumination |
US7425070B2 (en) * | 2005-05-13 | 2008-09-16 | Microsoft Corporation | Three-dimensional (3D) image projection |
-
2008
- 2008-04-17 WO PCT/US2008/060658 patent/WO2009011946A1/en active Application Filing
- 2008-04-17 US US12/105,120 patent/US8432436B2/en active Active
-
2013
- 2013-03-25 US US13/850,229 patent/US20140327748A1/en not_active Abandoned
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5365370A (en) * | 1993-06-10 | 1994-11-15 | Hudgins J Stephen | Three dimensional viewing illusion with 2D display |
US6744569B2 (en) * | 2001-06-19 | 2004-06-01 | Genex Technologies, Inc | Method and apparatus for omnidirectional three dimensional imaging |
US20070146358A1 (en) * | 2004-01-09 | 2007-06-28 | Koninklijke Philips Electronic, N.V. | Three-dimensional display |
US20070121182A1 (en) * | 2005-09-29 | 2007-05-31 | Rieko Fukushima | Multi-viewpoint image generation apparatus, multi-viewpoint image generation method, and multi-viewpoint image generation program |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9927628B2 (en) | 2014-07-18 | 2018-03-27 | National Institute Of Information And Communications Technology | Image display apparatus |
US10585393B2 (en) | 2015-08-28 | 2020-03-10 | Electronics And Telecommunications Research Institute | Apparatus for holographic display |
US10134190B2 (en) | 2016-06-14 | 2018-11-20 | Microsoft Technology Licensing, Llc | User-height-based rendering system for augmented reality objects |
Also Published As
Publication number | Publication date |
---|---|
US8432436B2 (en) | 2013-04-30 |
WO2009011946A1 (en) | 2009-01-22 |
US20080297593A1 (en) | 2008-12-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8432436B2 (en) | Rendering for an interactive 360 degree light field display | |
Jones et al. | Rendering for an interactive 360 light field display | |
US8087784B2 (en) | Systems and methods for displaying three-dimensional images | |
JP6489482B2 (en) | System and method for generating three-dimensional image media | |
US7385600B2 (en) | Three dimensional display method, system and apparatus | |
KR102491749B1 (en) | Panoramic 3D Imaging System | |
US10078228B2 (en) | Three-dimensional imaging system | |
JPH07306955A (en) | Method and system for generation of three-dimensional illusion | |
JP2008278087A (en) | Display device | |
KR20090045938A (en) | Three-dimensional display system | |
Luo et al. | Pepper's Cone: An Inexpensive Do-It-Yourself 3D Display | |
CN109769111A (en) | Image display method, device, system, storage medium and processor | |
RU2718777C2 (en) | Volumetric display | |
US10200675B1 (en) | Omni-directional stereo system | |
Bourke | Low Cost Projection Environment for Immersive Gaming. | |
RU2526901C1 (en) | Three-dimensional display and method of forming three-dimensional images | |
Collender | 3-D television, movies and computer graphics without glasses | |
Tanaka et al. | A method for the real-time construction of a full parallax light field | |
Otsuka et al. | Transpost: 360 deg-viewable three-dimensional display system | |
WO2023035965A1 (en) | Imaging apparatus, method and device | |
US20210306611A1 (en) | Multiview Image Capture and Display System | |
CA2532937C (en) | Three dimensional display method, system and apparatus | |
WO2005009051A1 (en) | Three dimensional display method, system and apparatus | |
Yano et al. | Full-parallax spherical light field display using mirror array | |
Nayar et al. | Projection Volumetric Display Using Passive Optical Scatterers |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: UNIVERSITY OF SOUTHERN CALIFORNIA, CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:DEBEVEC, PAUL E.;JONES, ANDREW;BOLAS, MARK;AND OTHERS;SIGNING DATES FROM 20080807 TO 20080817;REEL/FRAME:030080/0354 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |