This PDF file contains the front matter associated with SPIE-IS&T Proceedings Volume 7237, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

Since operation microscopes have been introduced to otology in the early 1950's, surgical reconstruction of middle ear function has been developed further to restore sound transmission with lowest loss possible. Today, research focuses on material properties for prosthetic replacement of middle ear ossicles and long-term outcome of reconstructive middle ear surgery. This study shows how stereoscopy is used to evaluate tissue-material interaction and its clinical impact for future restoration of middle ear function.

We conducted a comparative study of different stereoscopic display modalities (head-mounted display, polarized projection, and multiview lenticular display) to evaluate their efficacy in supporting manipulation and understanding of 3D content, specifically, in the context of neurosurgical visualization. Our study was intended to quantify the differences in resulting task performance between these choices of display technology. The experimental configuration involved a segmented brain vasculature and a simulated tumor. Subjects were asked to manipulate the vasculature and a pen-like virtual probe in order to define a vessel-free path from cortical surface to the targeted tumor. Because of the anatomical complexity, defining such a path can be a challenging task. To evaluate the system, we quantified performance differences under three different stereoscopic viewing conditions. Our results indicate that, on average, participants achieved best performance using polarized projection, and worst with the multiview lenticular display. These quantitative measurements were further reinforced by the subjects' responses to our post-test questionnaire regarding personal preferences.

This paper reports on the results of a study investigating the benefits of using an autostereoscopic display in the training targeting process of the Germain Air Force. The study examined how stereoscopic 3D visualizations can help to improve flight path planning and the preparation of a mission in general. An autostereoscopic display was used because it allows the operator to perceive the stereoscopic images without shutter glasses which facilitates the integration into a workplace with conventional 2D monitors and arbitrary lighting conditions.

In this paper, we report on the development of a 3D vision system consisting of a flat panel stereoscopic display and auto-converging stereo camera and an assessment of the system's use for robotic driving, manipulation, and surveillance operations. The 3D vision system was integrated onto a Talon Robot and Operator Control Unit (OCU) such that direct comparisons of the performance of a number of test subjects using 2D and 3D vision systems were possible. A number of representative scenarios were developed to determine which tasks benefited most from the added depth perception and to understand when the 3D vision system hindered understanding of the scene. Two tests were conducted at Fort Leonard Wood, MO with noncommissioned officers ranked Staff Sergeant and Sergeant First Class. The scenarios; the test planning, approach and protocols; the data analysis; and the resulting performance assessment of the 3D vision system are reported.

Modern computational science poses two challenges for scientific visualization: managing the size of resulting datasets and extracting maximum knowledge from them. While our team attacks the first problem by implementing parallel visualization algorithms on supercomputing architectures at vast scale, we are experimenting with autostereoscopic display technology to aid scientists in the second challenge. We are building a visualization framework connecting parallel visualization algorithms running on one of the world's most powerful supercomputers with high-quality autostereo display systems. This paper is a case study of the development of an end-to-end solution that couples scalable volume rendering on thousands of supercomputer cores to the scientists' interaction with autostereo volume rendering at their desktops and larger display spaces. We discuss modifications to our volume rendering algorithm to produce perspective stereo images, their transport from supercomputer to display system, and the scientists' 3D interactions. A lightweight display client software architecture supports a variety of monoscopic and autostereoscopic display technologies through a flexible configuration framework. This case study provides a foundation that future research can build upon in order to examine how autostereo immersion in scientific data can improve understanding and perhaps enable new discoveries.

From a signal processing perspective, we examine the main factors defining the visual quality of autostereoscopic 3-D displays, which are beginning to reproduce the plenoptic function with increasing accuracy. We propose using intuitive visual tools and ray-tracing simulations to gain insight into the signal processing aspects, and we demonstrate the advantages of analyzing what we call mixed spatial-angular spaces. With this approach we are able to intuitively demonstrate some basic limitations of displays using anisotropic diffusers or lens arrays. Furthermore, we propose new schemes for improved performance.

We developed a flat panel display having a slanted subpixel arrangement suitable for a multi-view display. A set of 3M × N subpixels (M × N subpixels for each R, G, and B color) is combined with one of the cylindrical lenses constituting a lenticular sheet to construct a three-dimensional pixel. Subpixels with the same color in each three-dimensional pixel have different horizontal positions, and these R, G, and B subpixels are repeated in the horizontal direction. In addition, ray-emitting areas of the subpixels in a three-dimensional pixel are continuous in the horizontal direction for each color. One vertical edge of one subpixel has the same horizontal position as the opposite vertical edge of one of the other subpixels of the same color subpixels. The crosstalk among viewpoints is theoretically zero and a three-dimensional image has no moire. When the horizontal pitch of subpixels is p, the width of a subpixel is 3p/N and the width of the black matrix region between subpixels is (1-3/N)p. This structure is suitable for providing a large number of viewpoints. A mobile three-dimensional display with 16 viewpoints and three-dimensional resolution of 256 × 192 was constructed. The screen size was 2.57 inches. A three-dimensional pixel was composed by 12 × 4 subpixels (M=4 and N=4). The horizontal pitch of subpixels was 17.0 μm and the horizontal width of the black matrix areas was 4.25 μm. The horizontal position of subpixels was shifted 9.56 μ;m between adjacent rows.

We propose a method to increase the viewing resolution of an autostereoscopic display without increasing the density of microlenses. Multiple projectors are used for the projection images to be focused and overlaid on a common plane in the air behind the microlens array. The multiple overlaid projection images yield multiple light spots inside the region of each elemental lenslet of the microlens array. This feature provides scalable high resolution images by increasing the number of projectors.

Advances in building high-performance camera arrays [1, 12] have opened the opportunity - and challenge - of using these devices for autostereoscopic display of live 3D content. Appropriate autostereo display requires calibration of these camera elements and those of the display facility for accurate placement (and perhaps resampling) of the acquired video stream. We present progress in exploiting a new approach to this calibration that capitalizes on high quality homographies between pairs of imagers to develop a global optimal solution delivering epipoles and fundamental matrices simultaneously for the entire system [2]. Adjustment of the determined camera models to deliver minimal vertical misalignment in an epipolar sense is used to permit ganged rectification of the separate streams for transitive positioning in the visual field. Individual homographies [6] are obtained for a projector array that presents the video on a holographically-diffused retroreflective surface for participant autostereo viewing. The camera model adjustment means vertical epipolar disparities of the captured signal are minimized, and the projector calibration means the display will retain these alignments despite projector pose variations. The projector calibration also permits arbitrary alignment shifts to accommodate focus-of-attention vengeance, should that information be available.

The recent resurgence of interest in the stereoscopic cinema and the increasing availability to the consumer of stereoscopic televisions and computer displays are leading broadcasters to consider, once again, the feasibility of stereoscopic broadcasting. High Definition Television is now widely deployed, and the R&D departments of broadcasters and consumer electronics manufacturers are starting to plan future enhancements to the experience of television. Improving the perception of depth via stereoscopy is a strong candidate technology. In this paper we will consider the challenges associated with the production, transmission and display of different forms of "three-dimensional" television. We will explore options available to a broadcaster wishing to start a 3D service using the technologies available at the present time, and consider how they could be improved to enable many more television programmes to be recorded and transmitted in a 3D-compatible form, paying particular attention to scenarios such as live broadcasting, where the workflows developed for the stereoscopic cinema are inapplicable. We will also consider the opportunities available for broadcasters to reach audiences with "three-dimensional" content via other media in the near future: for example, distributing content via the existing stereoscopic cinema network, or over the Internet to owners of stereoscopic computer displays.

Modern stereoscopic 3D (S3D) cinematic production relies heavily on digital tools and pipelines. Although technically possible, avoiding digital processing of the images is not only unlikely, but will probably yield worse results, since digital tools and pipelines inherently solve many of the issues found in traditional S3D film production and present the opportunity to correct for many others. All computer graphics (CG) tools are based on an implicit or explicit camera model. The model is even more overt in S3D productions and presents more complex implications and ramifications throughout the production pipeline. In this paper, the impact of the camera model on various processes in a production pipeline will be explored and some methods are proposed to improve efficiency with a savvy generation, recording, transport and use of metadata.

The lenticular raster system for 3D movies non-glasses show designed by NIKFI demonstrated commercially in Moscow in the 40'st of the last century. Essential lack of this method was narrow individual viewing zone as only two images on the film used. To solve this problem, we propose to use digital video projective system with modular principle of its design. Increase of the general number of the pixels forming the stereo image is reached by using of more than one projector. The modular projection autostereoscopic system for demonstration of the 3D movies includes diffuser screen; lenticular plate located in front of the screen; projective system consisted from several projectors and the block of parallax panoramogram fragments creator. By means of this block the parallax panoramogram is broken into fragments which quantity corresponds to number of projectors. For the large dimension lenticular screen making rectangular fragments of inclined raster were joined in a uniform leaf. To obtain the needed focal distance of the screen lenses we used immersion - aqueous solution of glycerin. The immersion also let essentially decrease visibility of fragments joints. An experimental prototype of the modular projection autostereoscopic system was created to validate proposed system.

Today there is a proliferation of stereoscopic 3D display devices, 3D content, and 3D enabled video games. As we in the S-3D community bring stereoscopic 3D to the home user we have a real opportunity of using stereoscopic 3D to bridge the gap between exciting immersive games and home movies. But to do this, we cannot limit ourselves to current conceptions of gaming and movies. We need, for example, to imagine a movie that is fully rendered using avatars in a stereoscopic game environment. Or perhaps to imagine a pervasive drama where viewers can play too and become an essential part of the drama - whether at home or on the go on a mobile platform. Stereoscopic 3D is the "glue" that will bind these video and movie concepts together. As users feel more immersed, the lines between current media will blur. This means that we have the opportunity to shape the way that we, as humans, view and interact with each other, our surroundings and our most fundamental art forms. The goal of this paper is to stimulate conversation and further development on expanding the current gaming and home theatre infrastructures to support greatly-enhanced experiential entertainment.

This paper presents an overview of the creative process and the technical challenges involved in the creation of the digital 3D presentation of the film Bolt, by Walt Disney Animation Studios.

The successful introduction of stereoscopic TV systems, such as Samsung's 3D Ready Plasma, requires high quality 3D content to be commercially available to the consumer. Console and PC games provide the most readily accessible source of high quality 3D content. This paper describes innovative developments in a generic, PC-based game driver architecture that addresses the two key issues affecting 3D gaming: quality and speed. At the heart of the quality issue are the same considerations that studios face producing stereoscopic renders from CG movies: how best to perform the mapping from a geometric CG environment into the stereoscopic display volume. The major difference being that for game drivers this mapping cannot be choreographed by hand but must be automatically calculated in real-time without significant impact on performance. Performance is a critical issue when dealing with gaming. Stereoscopic gaming has traditionally meant rendering the scene twice with the associated performance overhead. An alternative approach is to render the scene from one virtual camera position and use information from the z-buffer to generate a stereo pair using Depth-Image-Based Rendering (DIBR). We analyze this trade-off in more detail and provide some results relating to both 3D image quality and render performance.

Existing stereoscopic imaging algorithms can create static stereoscopic images with perceived depth control function to ensure a compelling 3D viewing experience without visual discomfort. However, current algorithms do not normally support standard Cinematic Storytelling techniques. These techniques, such as object movement, camera motion, and zooming, can result in dynamic scene depth change within and between a series of frames (shots) in stereoscopic cinematography. In this study, we empirically evaluate the following three types of stereoscopic imaging approaches that aim to address this problem. (1) Real-Eye Configuration: set camera separation equal to the nominal human eye interpupillary distance. The perceived depth on the display is identical to the scene depth without any distortion. (2) Mapping Algorithm: map the scene depth to a predefined range on the display to avoid excessive perceived depth. A new method that dynamically adjusts the depth mapping from scene space to display space is presented in addition to an existing fixed depth mapping method. (3) Depth of Field Simulation: apply Depth of Field (DOF) blur effect to stereoscopic images. Only objects that are inside the DOF are viewed in full sharpness. Objects that are far away from the focus plane are blurred. We performed a human-based trial using the ITU-R BT.500-11 Recommendation to compare the depth quality of stereoscopic video sequences generated by the above-mentioned imaging methods. Our results indicate that viewers' practical 3D viewing volumes are different for individual stereoscopic displays and viewers can cope with much larger perceived depth range in viewing stereoscopic cinematography in comparison to static stereoscopic images. Our new dynamic depth mapping method does have an advantage over the fixed depth mapping method in controlling stereo depth perception. The DOF blur effect does not provide the expected improvement for perceived depth quality control in 3D cinematography. We anticipate the results will be of particular interest to 3D filmmaking and real time computer games.

Stereoscopic imagery, by the fact that each eye requires a different image, defies mass marketing. Many have tried various methods over the last 150 years, with varying degrees of success. We will discuss the history of processes, and cover pros and cons of various methods.

Some people report visual discomfort when watching 3D displays. For both the objective measurement of visual fatigue and the subjective measurement of visual discomfort, we would like to arrive at general indicators that are easy to apply in perception experiments. Previous research yielded contradictory results concerning such indicators. We hypothesize two potential causes for this: 1) not all clinical tests are equally appropriate to evaluate the effect of stereoscopic viewing on visual fatigue, and 2) there is a natural variation in susceptibility to visual fatigue amongst people with normal vision. To verify these hypotheses, we designed an experiment, consisting of two parts. Firstly, an optometric screening was used to differentiate participants in susceptibility to visual fatigue. Secondly, in a 2×2 within-subjects design (2D vs 3D and two-view vs nine-view display), a questionnaire and eight optometric tests (i.e. binocular acuity, fixation disparity with and without fusion lock, heterophoria, convergent and divergent fusion, vergence facility and accommodation response) were administered before and immediately after a reading task. Results revealed that participants found to be more susceptible to visual fatigue during screening showed a clinically meaningful increase in fusion amplitude after having viewed 3D stimuli. Two questionnaire items (i.e., pain and irritation) were significantly affected by the participants' susceptibility, while two other items (i.e., double vision and sharpness) were scored differently between 2D and 3D for all participants. Our results suggest that a combination of fusion range measurements and self-report is appropriate for evaluating visual fatigue related to 3D displays.

In many application domains the analysis of aerial or satellite images plays an important role. The use of stereoscopic display technologies can enhance the image analyst's ability to detect or to identify certain objects of interest, which results in a higher performance. Changing image acquisition from analog to digital techniques entailed the change of stereoscopic visualisation techniques. Recently different kinds of digital stereoscopic display techniques with affordable prices have appeared on the market. At Fraunhofer IITB usability tests were carried out to find out (1) with which kind of these commercially available stereoscopic display techniques image analysts achieve the best performance and (2) which of these techniques achieve a high acceptance. First, image analysts were interviewed to define typical image analysis tasks which were expected to be solved with a higher performance using stereoscopic display techniques. Next, observer experiments were carried out whereby image analysts had to solve defined tasks with different visualization techniques. Based on the experimental results (performance parameters and qualitative subjective evaluations of the used display techniques) two of the examined stereoscopic display technologies were found to be very good and appropriate.

Humans actively explore their visual environment by moving their eyes. Precise coordination of the eyes during visual scanning underlies the experience of a unified perceptual representation and is important for the perception of depth. We report data from three psychological experiments investigating human binocular coordination during visual processing of stereoscopic stimuli.In the first experiment participants were required to read sentences that contained a stereoscopically presented target word. Half of the word was presented exclusively to one eye and half exclusively to the other eye. Eye movements were recorded and showed that saccadic targeting was uninfluenced by the stereoscopic presentation, strongly suggesting that complementary retinal stimuli are perceived as a single, unified input prior to saccade initiation. In a second eye movement experiment we presented words stereoscopically to measure Panum's Fusional Area for linguistic stimuli. In the final experiment we compared binocular coordination during saccades between simple dot stimuli under 2D, stereoscopic 3D and real 3D viewing conditions. Results showed that depth appropriate vergence movements were made during saccades and fixations to real 3D stimuli, but only during fixations on stereoscopic 3D stimuli. 2D stimuli did not induce depth vergence movements. Together, these experiments indicate that stereoscopic visual stimuli are fused when they fall within Panum's Fusional Area, and that saccade metrics are computed on the basis of a unified percept. Also, there is sensitivity to non-foveal retinal disparity in real 3D stimuli, but not in stereoscopic 3D stimuli, and the system responsible for binocular coordination responds to this during saccades as well as fixations.

Liquid Crystal Displays (LCD) are now a popular display technology for consumer television applications. Our previous research has shown that conventional LCD computer monitors are not well suited to time-sequential stereoscopic visualization due to the scanning image update method, the hold-type operation of LCDs, and in some cases slow pixel response rate. Recently some new technologies are being used in LCD TVs to improve 2D motion reproduction - such as black frame insertion and 100/120Hz capability. This paper reports on the testing of a selection of recent LCD TVs to investigate their compatibility with the time-sequential stereoscopic display method - particularly investigating new display technologies. Aspects considered in this investigation include image update method, pixel response rate, maximum input frame rate, backlight operation, frame rate up-conversion technique, synchronization, etc. A more advanced Matlab program was also developed as part of this study to simulate and characterize 3D compatibility and calculate the crosstalk present on each display. The results of the project show that black frame insertion does improve 3D compatibility of LCDs but not to a sufficient level to produce good 3D results. Unfortunately 100/120Hz operation of the tested LCD did not improve 3D compatibility compared to the LCD monitors tested previously.

A color holographic reconstruction technique by using three light emitting diodes (LEDs) is described. Reflective type phase-only spatial light modulators (SLMs) are used since they are suitable for in-line phase holograms. Gerchberg-Saxton iterative algorithm is used for computing phase holograms. Three phase holograms are calculated separately for red, green and blue colors, for a color reconstruction, and separately loaded to corresponding SLMs. Three LEDs are used for illuminating those phase holograms and reconstructions are combined and captured. Experimental results are satisfactory.

We developed two types of a new structured 2D/3D switchable Stereoscopic LCD display by making some changes to the structure of the conventional TFT-LCD. This new display can show both 2D image and 3D image, and has no limitations on neither viewing distance nor viewing angle. Thus, it is possible for multiple persons to view 3D image on this display simultaneously, and it is expected to be used as a 2D/3D switchable LCD panel in the next-generation Flat Panel LCD TV.

Context: The idea behind stereoscopic displays is to create the illusion of depth and this concept could have many practical applications. A common spatial ability test involves mental rotation. Therefore a mental rotation task should be easier if being undertaken on a stereoscopic screen. Aim: The aim of this project is to evaluate stereoscopic displays (3D screen) and to assess whether they are better for performing a certain task than over a 2D display. A secondary aim was to perform a similar study but replicating the conditions of using a stereoscopic mobile phone screen. Method: We devised a spatial ability test involving a mental rotation task that participants were asked to complete on either a 3D or 2D screen. We also design a similar task to simulate the experience on a stereoscopic cell phone. The participants' error rate and response times were recorded. Using statistical analysis, we then compared the error rate and response times of the groups to see if there were any significant differences. Results: We found that the participants got better scores if they were doing the task on a stereoscopic screen as opposed to a 2D screen. However there was no statistically significant difference in the time it took them to complete the task. We also found similar results for 3D cell phone display condition. Conclusions: The results show that the extra depth information given by a stereoscopic display makes it easier to mentally rotate a shape as depth cues are readily available. These results could have many useful implications to certain industries.

In stereo displays, binocular disparity creates a striking impression of depth. However, such displays present focus cues-blur and accommodation-that specify a different depth than disparity, thereby causing a conflict. This conflict causes several problems including misperception of the 3D layout, difficulty fusing binocular images, and visual fatigue. To address these problems, we developed a display that preserves the advantages of conventional stereo displays, while presenting correct or nearly correct focus cues. In our new stereo display each eye views a display through a lens that switches between four focal distances at very high rate. The switches are synchronized to the display, so focal distance and the distance being simulated on the display are consistent or nearly consistent with one another. Focus cues for points in--between the four focal planes are simulated by using a depth-weighted blending technique. We will describe the design of the new display, discuss the retinal images it forms under various conditions, and describe an experiment that illustrates the effectiveness of the display in maximizing visual performance while minimizing visual fatigue.

Large real-time holographic displays with full color are feasible with SeeReal's new approach to holography and today's technology. The display provides the information about the 3D scene in a viewing window at each observer eye. A tracking system always locates the viewing windows at the observer eyes. This combination of diffractive and refractive optics leads to a significant reduction of required display resolution and computation effort and enables holographic displays for wide-spread consumer applications. We tested our approach with two 20 inch prototypes that use two alternatives to achieve full color. One prototype uses color filters and interlaced holograms to generate the colors simultaneously. The other prototype generates the colors sequentially. In this paper we review our technology briefly, explain the two alternatives to full color and discuss the next steps toward a consumer product.

In this paper, a high-definition integral floating display is implemented. Integral floating display is composed of an integral imaging system and a floating lens. The integral imaging system consists of a two-dimensional (2D) display and a lens array. In this paper, we substituted multiple spatial light modulators (SLMs) for a 2D display to acquire higher definition. Unlike conventional integral floating display, there is space between displaying regions of SLMs. Therefore, SLMs should be carefully aligned to provide continuous viewing region and seamless image. The implementation of the system is explained and three-dimensional (3D) image displayed by the system is represented.

In this contribution, we will present a 3.5" 3D QVGA display that uses a highly-efficient, patterned and controllable OLED backlight. Several major achievements are included in this technology demonstrator, like large-area OLED backlight, highly-efficient and fast-response OLED top-emitter, striped patterned backlight, and individual electronic driving for adaptive backlight control. A 3D mobile display application has been successfully demonstrated.

Investigated H1-H2 transfer analogue and digitally printed reflection holograms suitability for autostereoscopic projection displays. Proved that reflection hologram, having part of replayed image in front of its surface may be used as autostereoscopic display. Assembled 3D streaming images projection device, incorporating digitally printed reflection hologram. Given recommendation for digitally printed holograms use as 3D projection screens.

A display system with lens arrays at the front of a high-resolution LCD has been known as a method to realize an autostereoscopic three-dimensional (3D) display. In these displays, a light ray overlap between neighboring parallax images affects the image quality. In this study, the overlap effects were investigated for the one-dimensional (horizontal parallax only) integral imaging (1D-II) method. We fabricated samples of 1D-II displays with different levels of light ray overlaps and evaluated the 3D image by subjective assessment. It is found that the 1D-II display utilizing the proper parallax overlaps can eliminate banding artifact and have good 3D image quality within the wide range of a viewing area.

Although a naked-eye 3D display is more convenient to watch for a viewer, so far and in the near future the image quality of a stereo display watched with special glasses is still much better than the former. e.g. the viewing angle, the crosstalk, the resolution, etc. While focusing on the glasses-type stereo display, the image performance of a time multiplexed shutter-glasses-type 3D display should be better than that of a spatial multiplexed polarization-encoded 3D display. Shutter-glasses-type 3D display was implemented many years ago by CRT. However, due to the generation supersedure the CRT was replaced by LCD, the shutter-glasses solution couldn't work for several years as a result of the long response time of LCD. Thanks to the development of over-drive technology, the response time of LCD is getting faster, and a 100-120Hz panel refresh rate is possible. Therefore, 3D game fans have a very good opportunity to watch full resolution, large viewing angle and low crosstalk stereo LCDs again. In this paper, a 120Hz LCD and an LED dynamic backlight to overcome the hold-type characteristic of an LCD are used to implement a time-multiplexed 3D display. A synchronization circuit is developed to connect the time scheme of the vertical sync. signal from the display card, the scanning backlight and the shutter glasses. The crosstalk under different scanning conditions is measured.

Auto-stereoscopic 3D displays offer presently the most attractive solution for entertainment and media consumption. Despite many studies devoted to this type of technology, efficient characterization methods are still missing. We present here an innovative optical method based on high angular resolution viewing angle measurements with Fourier optics instrument. This type of instrument allows measuring the full viewing angle aperture of the display very rapidly and accurately. The system used in the study presents a very high angular resolution below 0.04 degree which is mandatory for this type of characterization. We can predict from the luminance or color viewing angle measurements of the different views of the 3D display what will be seen by an observer at any position in front of the display. Quality criteria are derived both for 3D and standard properties at any observer position and Qualified Stereo Viewing Space (QSVS) is determined. The use of viewing angle measurements at different locations on the display surface during the observer computation gives more realistic estimation of QSVS and ensures its validity for the entire display surface. Optimum viewing position, viewing freedom, color shifts and standard parameters are also quantified. Simulation of the moire issues can be made leading to a better understanding of their origin.

In integral imaging, lens arrays are used to capture the image of the object and display the three-dimensional (3-D) image. In principle, the 3-D image is reconstructed at the position where the object was. We have hitherto proposed a method for controlling the depth position of the reconstructed image by applying numerical processing to the captured image information. First, the rays from the object are regenerated by numerical processing by using information captured from the actual object together with a first virtual lens array. Next, the regenerated rays are used to generate 3-D information corresponding to a prescribed depth position by arranging a second virtual lens array. In this paper, we clarify the spatial frequency relationship between the object and the depthcontrolled reconstructed image, and we propose filter characteristics that can be used to avoid aliasing. We also report on experiments in which we confirm the effectiveness of the proposed filter.

Stereoscopic video delivers depth perception to users contrary to 2-dimenstional video. Therefore, we need to develop a new video quality assessment model for stereoscopic video. In this paper, we propose a new method for objective assessment of stereoscopic video. The proposed method detects blocking artifacts and degradation in edge regions such as in conventional video quality assessment model. In addition, it detects video quality difference between views using depth information. We performed subjective evaluation of stereoscopic video to verify the performance of the proposed method, and we confirmed that the proposed algorithm is superior to PSNR in respect to correlation with the subjective evaluation.

A vastly growing number of productions from the entertainment industry are aiming at 3D movie theatres. These productions use a two-view format, primarily intended for eye-wear assisted viewing in a well defined environment. To get this 3D content into the home environment, where a large variety of 3D viewing conditions exists (e.g different display sizes, display types, viewing distance), we need a flexible 3D format that can adjust the depth effect. Such a format is the image plus depth format in which a video frame is enriched with depth information of all pixels in the video. This format can be extended with an additional layer for occluded video and associated depth, that contains what is behind objects in the video. To produce 3D content in this extended format, one has to deduce what is behind objects. There are various axes along which this occluded data can be obtained. This paper presents a method to automatically detect and fill the occluded areas exploiting the temporal axis. To get visually pleasing results, it is of utmost importance to make the inpainting globally consistent. To do so, we start by analyzing data along the temporal axis and compute a confidence for each pixel. Then pixels from the future and the past that are not visible in the current frame are weighted and accumulated based on computed confidences. These results are then fed to a generic multi-source framework that computes the occlusion layer based on the available confidences and occlusion data.

In this paper we present experiments and results pertaining to the perception of depth in stereoscopic viewing of synthetic imagery. In computer animation, typical synthetic imagery is highly textured and uses stylized illumination of abstracted material models by abstracted light source models. While there have been numerous studies concerning stereoscopic capabilities, conventions for staging and cinematography in stereoscopic movies have not yet been well-established. Our long-term goal is to measure the effectiveness of various cinematography techniques on the human visual system in a theatrical viewing environment. We would like to identify the elements of stereoscopic cinema that are important in terms of enhancing the viewer's understanding of a scene as well as providing guidelines for the cinematographer relating to storytelling. In these experiments we isolated stereoscopic effects by eliminating as many other visual cues as is reasonable. In particular, we aim to empirically determine what types of movement in synthetic imagery affect the perceptual depth sensing capabilities of our viewers. Using synthetic imagery, we created several viewing scenarios in which the viewer is asked to locate a target object's depth in a simple environment. The scenarios were specifically designed to compare the effectiveness of stereo viewing, camera movement, and object motion in aiding depth perception. Data were collected showing the error between the choice of the user and the actual depth value, and patterns were identified that relate the test variables to the viewer's perceptual depth accuracy in our theatrical viewing environment.

In this paper, we propose a method for compressive acquisition of Ray-Space. Briefly speaking, incomplete data which directly captured by a specific device is transformed to full information by Radon transform. Ray-Space, which represents 3D images, describes position and direction of rays on reference plane in real space. Ray-Space has information of many rays. In conventional acquisition of Ray-Space, multiple cameras are used and 1 pixel on a camera captures 1 ray. Thus we need many pixels and we must capture the large amount of data. However Ray-Space has redundancy because Ray-Space consists of set of lines which depend on the depth of objects. We use the Radon transform to exploit this redundancy. The Radon transform is set of projection data along different directions. The Radon transform of Ray-Space show uneven distribution. Thus Ray-Space can be reconstructed from projection data in limited range by the inverse Radon transform. Capturing the part of projection data correspond to capturing sums of several rays by 1 pixel. A sum of several rays means a sum of brightness of rays. In this paper, we have simulated reconstruction of Ray-Space projection data which was computed by the Radon Transform of Ray-Space. This experiment showed that Ray-Space could be reconstructed from the parts of projection data. As a result, using fewer pixels than rays, we could reduce the amount of data to reconstruct Ray-Space.

We researched the influence of 3-D perception on confidence in decision-making. How does task complexity influence our self-confidence when hard decisions have to be taken? These questions are of high importance in various fields, including minimally invasive surgery and eyewitness recognition, since an incorrect decision may have severe consequences. This behavioral study compared 2-D and 3-D environments for the following parameters: correctness of results, level of confidence, confidence bias and interaction between task complexity and level of confidence. A group of young adults (N=41) were presented with a visual perceptual task in both the 2-D and 3-D environment. Results indicated significantly higher decision-making confidence levels in the 3-D environment, relative to the correctness of the decision. Also indicated were lower confidence bias in the 3-D environment and a significant interaction between self-confidence and task complexity between the 2-D/3-D environments.

Stereoscopic displays are increasingly used for computer-aided design. The aim is to make virtual prototypes to avoid building real ones, so that time, money and raw materials are saved. But do we really know whether virtual displays render the objects in a realistic way to potential users? In this study, we have performed several experiments in which we compare two virtual shapes to their equivalent in the real world, each of these aiming at a specific issue by a comparison: First, we performed some perception tests to evaluate the importance of head tracking to evaluate if it is better to concentrate our efforts on stereoscopic vision; Second, we have studied the effects of interpupillary distance; Third, we studied the effects of the position of the main object in comparison with the screen. Two different tests are used, the first one using a well-known shape (a sphere) and the second one using an irregular shape but with almost the same colour and dimension. These two tests allow us to determine if symmetry is important in their perception. We show that head tracking has a more important effect on shape perception than stereoscopic vision, especially on depth perception because the subject is able to move around the scene. The study also shows that an object between the subject and the screen is perceived better than an object which is on the screen, even if the latter is better for the eye strain.

The study of humans' perceptual experiences in Virtual Environments (VEs) plays an essential role in Virtual Reality (VR) research field. In particular, in the last years several researches were proposed regarding the problem if depth and distance are perceived in VEs as they are perceived in Real Environments (REs), and possibily what conditions affect a non correct estimation by the observers. This problem is very relevant in order to use VR as a supporting tool in fields where correct perception of space and distance is vital, like e.g. the training of personnel in dangerous environments. Many theories have been suggested regarding the combination and relation between different depth cues; unfortunately, no conclusive answer has been proposed. However, a common conclusion between all the experiments is that observers underestimate long distances in VEs. Although the causes of this phenomenon are still uncertain, it's reasonable to speculate that something must differ in the way distance and depth are extracted and processed between the RE and the VE. Moreover, it is worth noting that very few works considered VR installations with large projection screen, covering a large field of view (FOV) in the observation process. In this paper, we aim at investigating depth perception in the Virtual Theater of the University of Milan, a VR installation characterized by a large semicylindrical screen that covers 120° of horizontal FOV. For its characteristics, the Virtual Theater represents an interesting and never considered test ground for psychophysical experiments regarding deph perception in VEs. We will present some preliminar perceptual matching experiments regarding the effect of shadows and reflections in the estimation of distances in VEs, and we will discuss the obtained results.

In this paper, we propose a depth map quality metric for three-dimensional videos which include stereoscopic videos and autostereoscopic videos. Recently, a number of researches have been done to figure out the relationship of perceptual quality and video impairment caused by various compression methods. However, we consider non-compression issues which are induced during acquisition and displaying. For instance, using multiple cameras structure may cause impairment such as misalignment. We demonstrate that the depth map can be a useful tool to find out the implied impairments. The proposed quality metrics using depth map are depth range, vertical misalignment, temporal consistency. The depth map is acquired by solving corresponding problems from stereoscopic video, widely known as disparity estimation. After disparity estimation, the proposed metrics are calculated and integrated into one value which indicates estimated visual fatigue based on the results of subjective assessment. We measure the correlation between objective quality metrics and subjective quality results to validate our metrics.

The ability to convert 2D video material to 3D would be extremely valuable for the 3D-TV industry. Such conversion might be achieved using depth maps extracted from the original 2D content. We previously demonstrated that surrogate depth maps with limited or imprecise depth information could be used to produce effective stereoscopic images. In the current study, we investigated whether gray intensity images associated with the Cr colour component of standard 2D-colour video sequences could be used effectively as surrogate depth maps. Colour component-based depth maps were extracted from ten video sequences and used to render images for the right-eye view. These were then combined with the original images for the left-eye view to form ten stereoscopic test sequences. A panel of viewers assessed the depth quality and the visual comfort of the synthesized test sequences and, for comparison, of monoscopic and camera-captured stereoscopic versions of the same sequences. The data showed that the ratings of depth quality for the synthesized test sequences were higher than those of the monoscopic versions, but lower than those of the camera-captured stereoscopic versions. For visual comfort, ratings were lower for the synthesized than for the monoscopic sequences but either equal to or higher than those of the camera-captured versions

In this paper, we address the handling of independently moving objects (IMOs) in automatic 2D to stereoscopic 3D conversion systems based on structure-from-motion (SfM) techniques. Exploiting the different viewing positions of a moving camera, these techniques yield excellent 3D results for static scene objects. However, the independent motion of any foreground object requires a separate conversion process. We propose a novel segmentation approach that estimates the occluded static background and segments the IMOs based on advanced change detection. The background estimation is achieved applying 2D registration and blending techniques, representing an approximation of the underlying scene geometry. The segmentation process itself uses anisotropic filtering applied on the difference image between original frame and the estimated background frame. In order to render the segmented objects into the automatically generated 3D scene properly, a small amount of user interaction will be necessary, e.g. an assignment of intra-object depth or the object's absolute z-position. Experiments show that the segmentation method achieves accurate mask results for a variety of scenes, similar to the masks obtained manually using state-of-the-art rotoscoping tools. Though, this work contributes to the extension of SfM-based automatic 3D conversion methods for the application on dynamic scenes.

Stereoscopic digital photography has become much more practical with the use of USB wired connections between a pair of Canon cameras using StereoData Maker software for precise synchronization. StereoPhoto Maker software is now used to automatically combine and align right and left image files to produce a stereo pair. Side by side images are saved as pairs and may be viewed using software that converts the images into the preferred viewing format at the time of display. Stereo images may be shared on the internet, displayed on computer monitors, autostereo displays, viewed on high definition 3D TVs, or projected for a group. Stereo photographers are now free to control composition using point and shoot settings, or are able to control shutter speed, aperture, focus, ISO, and zoom. The quality of the output depends on the developed skills of the photographer as well as their understanding of the software, human vision and the geometry they choose for their cameras and subjects. Observers of digital stereo images can zoom in for greater detail and scroll across large panoramic fields with a few keystrokes. The art, science, and methods of taking, creating and viewing digital stereo photos are presented in a historic and developmental context in this paper.

This paper presents a stereoscopic contents authoring system that covers the creation and editing of stereoscopic multimedia contents for the 3D DMB (Digital Multimedia Broadcasting) data services. The main concept of 3D DMB data service is that, instead of full 3D video, partial stereoscopic objects (stereoscopic JPEG, PNG and MNG) are stereoscopically displayed on the 2D background video plane. In order to provide stereoscopic objects, we design and implement a 3D DMB content authoring system which provides the convenient and straightforward contents creation and editing functionalities. For the creation of stereoscopic contents, we mainly focused on two methods: CG (Computer Graphics) based creation and real image based creation. In the CG based creation scenario where the generated CG data from the conventional MAYA or 3DS MAX tool is rendered to generate the stereoscopic images by applying the suitable disparity and camera parameters, we use X-file for the direct conversion to stereoscopic objects, so called 3D DMB objects. In the case of real image based creation, the chroma-key method is applied to real video sequences to acquire the alpha-mapped images which are in turn directly converted to stereoscopic objects. The stereoscopic content editing module includes the timeline editor for both the stereoscopic video and stereoscopic objects. For the verification of created stereoscopic contents, we implemented the content verification module to verify and modify the contents by adjusting the disparity. The proposed system will leverage the power of stereoscopic contents creation for mobile 3D data service especially targeted for T-DMB with the capabilities of CG and real image based contents creation, timeline editing and content verification.

High-quality stereoscopic image content must be viewable in a variety of visual environments, from 3-D theaters to 3-D mobile devices. Stereoscopic effects, however, are affected by screen size, viewing distance, and other parameters. In this study, the authors focus on the stereoscopic image quality experience of viewing 3-D content on a mobile device in order to compare it with that of viewing 3-D content on a large screen. The stereoscopic image quality experience was evaluated using Interpretation Based Quality (IBQ) methodology, which combines existing approaches to image quality evaluation, such as the paired comparison and interview, and assesses the viewer experience using both quantitative and qualitative data. Five stereoscopic images were used in the experiment. The results of the experiment suggest that the discomfort felt while viewing stereoscopic images on a 3-D mobile device arise from not only visual fatigue but also the effects of the smaller screen size. The study also revealed the types of stereoscopic images that are suitable for viewing on 3-D mobile devices.

We identify, categorize and simulate artifacts which might occur during delivery stereoscopic video to mobile devices. We consider the stages of 3D video delivery dataflow: content creation, conversion to the desired format (multiview or source-plus-depth), coding/decoding, transmission, and visualization on 3D display. Human 3D vision works by assessing various depth cues - accommodation, binocular depth cues, pictorial cues and motion parallax. As a consequence any artifact which modifies these cues impairs the quality of a 3D scene. The perceptibility of each artifact can be estimated through subjective tests. The material for such tests needs to contain various artifacts with different amounts of impairment. We present a system for simulation of these artifacts. The artifacts are organized in groups with similar origins, and each group is simulated by a block in a simulation channel. The channel introduces the following groups of artifacts: sensor limitations, geometric distortions caused by camera optics, spatial and temporal misalignments between video channels, spatial and temporal artifacts caused by coding, transmission losses, and visualization artifacts. For the case of source-plus-depth representation, artifacts caused by format conversion are added as well.

To be able to accommodate different viewing conditions (e.g. different display sizes, display types, viewing distance), a 3D format is required where the depth effect can be adjusted. Within MPEG, the image-and-depth format has been standardized. This format is bandwidth efficient, highly flexible and therefore well suited to introduce 3D content into for example the home environment. Extensions to this format have been proposed to enable occlusion handling, by introducing more image-and-depth layers, containing image data located behind the foreground objects, to enable looking around such foreground objects. In the presence of such a multi-layer representation, a next extension is to add transparency information to the layers, allowing for alpha matting (offering significant gains in rendering quality) and inclusion of semi-transparent objects in the scene. Due to the multi-layer nature, it is then still possible to tune the amount of depth. In this paper, we report on our design choices to arrive at the Declipse 2 format, by going through the whole video chain, from content creation, to looking at the effects of video compression and finally real-time 3D rendering and display of such multi-layer content with transparency information, showing the feasibility of the approach under practical conditions.

Autostereoscopic displays, allows one to see in 3D without glasses it requires at least 2 views and is often to maximize user comfort. However multi-view capture is too complex for more than two cameras. Furthermore there is no interoperability between 3D displays because each manufacturer has his specific input format. Thus we propose an efficient automatic conversion from stereoscopic views to N views for a rendering on various autostereoscopic screens. We describe the 2 main steps achieving automatic conversion from 2 to N views: disparity map estimation and view interpolation. We use graph cut algorithm to compute the disparity map with occlusions detection. As the quality of the disparity map is essential for interpolation we have developed an efficient management of occlusions through morphological operation, outliers erasing and padding with the smallest disparity. We thus reconstruct details even for concave objects. We also apply spatial regularization preserving borders and temporal smoothing with motion compensation. Those processes improve considerably the quality of the depths map videos. Virtual views are then generated through view morphing based on disparity to avoid 3D manipulation of data. Occlusion areas are managed and irregular morphing pixel locations are resampled with a B-spline filter adapted to view interpolation.

In this paper, we present the depth value accuracy requirements for novel view synthesis in 3DTV and our approach for depth camera based on hybrid method of structured light and active stereo matching. We first review the background on the use of multiview video plus depth(MVD) representation for future 3DTV system, depth acquisition methods and tools for real-time and non-real-time dense depth map acquisition in terms of performance and limitations. Then we use a simple analysis model for novel view synthesis based on DIBR to draw depth value accuracy requirements. Finally we present our approach on depth camera development along with future directions and huddles in developing real-time depth camera usable for 3DTV content production.

We developed a stereo-photogrammetry camera mounted system and assess its application as an imaging modality suitable for image-guided radiation therapy (IGRT). 2D clinical photography of visible disease is routinely employed to document tumor response and the side effects of treatment interventions. IGRT is the process of incorporating new patient imaging data into the radiation treatment course in order to make adjustments to daily treatment delivery parameters. Our system, which combines the two, resides within the head of a linear accelerator treatment machine, which lends its computer-controlled precision to acquire stereo pairs with millimeter accuracy. Our IGRT workflow is contained in our analysis.

In depth image based rendering, video sequences and their associated depth maps are used to render new camera viewpoints for stereoscopic applications. In this study, we examined the effect of temporal downsampling of the depth maps on stereoscopic depth quality and visual comfort. The depth maps of four eight-second video sequences were temporally downsampled by dropping all frames, except the first, for every 2, 4, or 8 consecutive frames. The dropped frames were then replaced by the retained frame. Test stereoscopic sequences were generated by using the original image sequences for the left-eye view and the rendered image sequences for the right-eye view. The downsampled versions were compared to a reference version with full depth maps that were not downsampled. Based on the data from 21 viewers, ratings of depth quality for the downsampled versions were lower. Importantly, ratings depended on the content characteristics of the stereoscopic video sequences. Results were similar for visual comfort, except that the differences in ratings between sequences were larger. The present results suggest that more processing, such as interpolation of depth maps, might be required to counter the negative effects of temporal downsampling, especially beyond a downsampling of two.

Binocular indirect ophthalmoscope (BIO) provides a wider view of fundus with stereopsis contrary to the direct one. Proposed system is composed of portable BIO and 3D viewing unit. The illumination unit of BIO utilized high flux LED as a light source, LED condensing lens cap for beam focusing, color filters and small lithium ion battery. In optics unit of BIO, beam splitter was used to distribute an examinee's fundus image both to examiner's eye and to CMOS camera module attached to device. Captured retinal video stream data from stereo camera modules were sent to PC through USB 2.0 connectivity. For 3D viewing, two video streams having parallax between them were aligned vertically and horizontally and made into side-by-side video stream for cross-eyed stereoscopy. And the data were converted into autostereoscopic video stream using vertical interlacing for stereoscopic LCD which has glass 3D filter attached to the front side of it. Our newly devised system presented the real-time 3-D view of fundus to assistants with less dizziness than cross-eyed stereoscopy. And the BIO showed good performance compared to conventional portable BIO (Spectra Plus, Keeler Limited, Windsor, UK).

Stereoscopic filming is roughly divided into two types: toed-in and parallel camera configurations. Both types have disadvantages: toed-in cameras cause keystone distortions, and parallel cameras cause image loss by shifting. In addition, it is difficult for inexperienced creators to understand the optimal camera settings and post-processing procedures, such as cross points and inter-camera distance, in both types. These factors hinder the creation of stereoscopic images. Therefore, the authors focused on improving usability in stereoscopic filming, constructed an experimental camera system, and examined semi-automatic camera configuration function in terms of viewing safety.

A novel autostereoscopic display system, the Integrated-Screen autostereoscopic display system, has been developed to substantially increase the total number of pixels on the screen, which in turn increase both the resolution and number of view-zones of the 3D display. In this system, a series of miniature projectors are arrayed and the projection images are tiled together seamlessly to form an image of ultra high resolution. For displaying 3D images, a lenticular screen with pre-designed tilted angle is used to distribute the pixels into the plural view-zones. In this paper, an Integrated-Screen autostereoscopic display system with 30" screen in diagonal and 15 view-zones is presented. The total resolution of the tiled image is 2930×2700, which is much higher than traditional Full HD display, and the resulted 3D resolution in each view-zone is 880×600.

It has been reported that even users of virtual environments and entertainment systems experience motion sickness. This visually induced motion sickness (VIMS) is known to be caused by sensory conflict, for instance, the disagreement between vergence and visual accommodation while viewing stereoscopic images. The simulator sickness questionnaire is a well-known tool that is used herein for verifying the occurrence of VIMS. We used the SSQ and also quantitatively measured head acceleration and sway of the center of gravity of the human body before and during the exposure to stereoscopic images on a head-mounted display. During the measurement, the subjects were instructed to maintain the Romberg posture for the first 60 s and a wide stance (with the midlines of heels 20 cm apart) for the next 60 s. We proposed a method to obtain stochastic differential equations (SDEs) as a mathematical model of the body sway on the basis of the stabilogram. While there are several minimal points of time-averaged potential function in the SDEs, the exposure decreases the gradient of the potential function. We have succeeded in estimating the decrease in the gradient of the potential function by using an index called sparse density.

We implemented the dense light field microscopy using the infinity corrected optical system. In the infinity corrected optical system, the three-dimensional specimen located around the focal plane of the objective is imaged at the intermediate plane by the combination of the objective and the tube lens. This intermediate image is again imaged by the micro lens array and captured by the CCD, providing the light field information. We analyzed geometrical structure of the dense light field microscope for infinity corrected optical system. From the analyzed results, we defined the characteristic and relationship of each component. Based on this result, we reconstructed various orthographic view images of the specimen from the captured light field, and also generated the depth slice images using the computational integral imaging reconstruction principle.

This paper presents a graphical software infrastructure for stereo display. It enables the development of low-cost/short development cycle stereo applications that are portable - not only across platforms, but across display types as well. Moreover, it allows not just images but entire GUI's (Graphics User Interface) to be displayed in stereo consistently across many platforms. Java Advanced Display Infrastructure for Stereo (JADIS) provides a common interface for displaying GUI components in stereo using either specialized stereo display hardware (e.g. liquid crystal shutter or polarized glasses) or anaglyph display (red/blue glasses) on standard computer displays. An application using this toolkit will work without modification in either environment, allowing stereo software to reach a wider audience (anaglyphs) without sacrificing high-quality display on dedicated hardware. JADIS has been released as Open Source and is available via the Open Channel foundation website^[1]. It has been integrated into several applications for stereo viewing and processing of data acquired by current and future NASA Mars surface missions (e.g. Mars Exploration Rover (MER), Phoenix Lander, Mars Science Laboratory (MSL)).

Display adaptation is one of the key factors for the success of visual services. For two-dimensional (2D) imaging, the display adaptation is generally accomplished by 2D image re-sampling (i.e., up-/down-sampling). However, when it comes to stereoscopic three-dimensional (S3D) images, 2D re-sampling methods are inadequate because additional consideration on the third dimension of depth is not incorporated. In this paper, we analyze S3D image resizing from two aspects: geometrical deformation and frequency-domain aliasing. A number of S3D displays are available in the market and they have various screen dimensions. As we have more varieties of the displays, efficient S3D image resizing is becoming more emphasized. We present the conditions for equi-proportional S3D image resizing using the model of the binocular vision; and propose an anti-aliasing filter for stereoscopic 3D image up/down-sampling.

A super multi-view (SMV) display and a high-density directional (HDD) display have both been proposed as natural three-dimensional displays that would solve the accommodation-vergence conflict and offer smooth motion parallax. Previously, a multiple projection system was employed in the construction of a HDD display. In this study, a multiple projection system is modified to serve as a SMV display. A lens is attached to the display screen. The screen lens images an aperture array in the multiple projection system at a fixed distance from the screen, and the image of the aperture array generates dense viewpoints. A 128-direction HDD display having ~QVGA resolution is employed to construct a 128-view SMV display. Fresnel lenses with focal lengths of 500 mm, 600 mm, 1,000 mm, 1,200 mm, and 1,500 mm were used as the screen lens. When the observation distance differs from the distance at which viewpoints are generated, a retinal image is formed by an assembly of plural parallax images. The dependency of image quality on observation distance is evaluated. Subjective evaluation shows that a Fresnel lens with a long focal length, i.e., 1,200 mm, gives better image quality for a wider observation distance range.

We present a simple depth estimation framework for 2D-to-3D media conversion. The perceptual depth information from monocular image is estimated by the optimal use of relative height cue, which is one of well-known depth recovery cues. The height depth cue is very common in photographic images. We propose a novel line tracing method and depth refinement filter as core of our depth estimation framework. The line tracing algorithm traces strong edge positions to generate an initial staircase depth map. The initial depth map is further improved by a recursive depth refinement filter. We present visual results from depth estimation and stereo image generation.

This paper proposes a new display device for switching between 3D and 2D images. A concave lens array is formed on the surface of a first panel, while a convex lens array is formed on the surface of a second panel. An image-plane of display, the first panel and the second panel are arranged in this order, and the concave lens and the convex lens face each other. When the two panels are brought into contact with each other, they show 2D images equivalent to that of one transparent plate. But when the separation interval between the panels is optimum, we see 3D images similar to that caused by lenticular lenses. This display is applied to a 3D and 2D switchable mobile display.

Flat 2D screens cannot display complex 3D structures without the usage of different slices of the 3D model. Volumetric displays like the "FELIX 3D-Displays" can solve the problem. They provide space-filling images and are characterized by "multi-viewer" and "all-round view" capabilities without requiring cumbersome goggles. In the past many scientists tried to develop similar 3D displays. Our paper includes an overview from 1912 up to today. During several years of investigations on swept volume displays within the "FELIX 3D-Projekt" we learned about some significant disadvantages of rotating screens, for example hidden zones. For this reason the FELIX-Team started investigations also in the area of static volume displays. Within three years of research on our 3D static volume display at a normal high school in Germany we were able to achieve considerable results despite minor funding resources within this non-commercial group. Core element of our setup is the display volume which consists of a cubic transparent material (crystal, glass, or polymers doped with special ions, mainly from the rare earth group or other fluorescent materials). We focused our investigations on one frequency, two step upconversion (OFTS-UC) and two frequency, two step upconversion (TFTSUC) with IR-Lasers as excitation source. Our main interest was both to find an appropriate material and an appropriate doping for the display volume. Early experiments were carried out with CaF2 and YLiF4 crystals doped with 0.5 mol% Er3+-ions which were excited in order to create a volumetric pixel (voxel). In addition to that the crystals are limited to a very small size which is the reason why we later investigated on heavy metal fluoride glasses which are easier to produce in large sizes. Currently we are using a ZBLAN glass belonging to the mentioned group and making it possible to increase both the display volume and the brightness of the images significantly. Although, our display is currently monochrome, it is possible to create an RGB-display. For the same reasons we started tests with polymers. We were able to achieve meaningful results which point out a new direction in the investigation on polymers. For the reasons described above, our new solid state device is one of modular design. The simplicity to change all components makes it possible to do experiments with different display volumes and lasers for every specific purpose of the display in a very effective way. The images can be drawn inside the display volume by acousto-optic, galvanometric or polygon mirror deflection units. We control our galvanometric deflection unit with a personal computer and a selfwritten software which makes it easier to handle the setup and makes interactivity possible. This setup makes it a powerful and flexible tool to keep track with the rapid technological progress of today and helped us to experience the disadvantages and the advantages of most of the possible deflection units in practice. These experiences are a main element in our paper and lead to some conclusions which will be of big importance in future display developments. Potential applications include imaging and computer aided design as well as scientific data visualization.

We present a new generation of liquid crystal shutters for active glasses, well suited to 3-D cinema current trends, involving triple flash regimes. Our technology uses a composite smectic C* liquid crystal mixture1. In this paper we focus on the electro-optical characterization of composite smectic-based shutters, and compare their performance with nematic ones, demonstrating their advantages for the new generation of 3-D cinema and more generally 3-D HDTV.

A multi-functional display is realized by utilizing optical processing based on spatial coding. The proposed display performs 2-D, 3-D, and peeping-prevention display. The display consists of two spatial light modulators of polarization-processing type. By use of optical encryption, the viewing zone is limited in three-dimensional space. By use of information sharing, stereoscopic images are provided. By use of modulation based on depth, depth-fused 3-D display is also realized. Without spatial modulation, conventional 2-D images are provided. Thus, the proposed display performs multi-functions without changing hardware configuration.

In this work, a new stereo image coding technique is proposed. The new approach integrates the coding of the residual image with the disparity map. The latter computed in the wavelet transform domain. The motivation behind using this transform is that it imitates some properties of the human visual system (HVS), particularly, the decomposition in the perspective canals. Therefore, using the wavelet transform allows for better perceptual image quality preservation. In order to estimate the disparity map, we used a quadtree segmentation in each wavelet frequency band. This segmentation has the advantage of minimizing the entropy. Dyadic squares in the subbands of target image that they are not matched with other in the reference image constitutes the residuals are coded by using an arithmetic codec. The obtained results are evaluated by using the SSIM and PSNR criteria.

We propose all-in-focus plane reconstruction based on computational integral imaging reconstruction. The depth of the object is detected by the pixel matching method with proper object mask, and computational integral imaging reconstruction is performed to get all in focus image. The pixel matching detects the depth of the single plane object by evaluating the difference between the collected rays. This pixel matching method can be extended to the multiple plane objects case by the use of the object masking. After each object is identified, the pixel matching is performed to one object by masking other objects. By repeating this process to all objects, the correct depth of the multiple objects can be detected. Computational integral imaging reconstruction is performed to all objects with detected depth values, resulting in all in focus image. From experimental and simulation results, it is confirmed that our

The three-dimensional (3D) displays provide a dramatic improvement of visual quality than the 2D displays do. The conversion of existing 2D videos to 3D videos is necessary for multimedia application. This paper presents a robust system to convert 2D videos to 3D videos. The main concepts are to extract the depth information from motion parallax of moving picture and to depth information from geometrical perspective in non-moving scene. In the first part, depthinduced motion information is reconstructed by motion vector to disparity mapping. By warping the consecutive video frames to parallel view angle with the current frame, the frame with suitable baseline is selected to generate depth using motion parallax information. However, video may not have the depth-induced motion information in every case. For scene without motion parallax, depth from geometrical perspective is applied to generate scene depth map. Scene depth map is assigned depending on the scene mode and analyzed line structure in the video. Combining these two depth cues, the stereo effect is enhanced and provide spectacular depth map. The depth map is then used to render the multi-view video for 3D display.

In this paper, we describe a method to combine two integral photography (IP) displays to represent a larger amount of depth while maintaining image quality. We adopt integral videography (IV), which can emit 4D light fields into real space to present full-parallax 3D videos. The two IVs are located at different depths on the same optical axis by using a beam splitter. We present some steps to enhance the quality of our 3D display. First, we remove positional displacements based on an adjustment between the LCD and the lens array of IV. The positional displacements include parallel, rotational, and depth displacement. Next, we strictly adjust the two IVs' positions to right positions. Adjusting geometrical positions is based on optical rotations and shifts. Finally, we run the hidden surface removal to hide surfaces of 3D objects on the background display to prevent viewers from seeing the distracting surface. In conclusion, our optically multilayered light field display is effective for enhancing the depth of field.

A real-time method for rendering integral photography (IP) that uses the extended fractional view technique is described. To make an IP image by using CG technology, hundreds of still pictures from different camera positions need to be taken, and the images have to be synthesized by using other software. This is because CG applications do not have a special rendering mode that is required for the extended fractional view approach when the directions of the rays are not uniform. Hence, considerable processing time is needed to synthesize one IP image using the method, which is not suitable for real-time applications such as games. Therefore, new high-speed rendering software was written using C++. It runs on a typical Windows PC. Its main function is to trace the path of each ray, which is emitted from each subpixel of a liquid crystal display (LCD) and refracted by a fly's eye lens. A subpixel is used instead of a pixel because a pixel on a color LCD is made up of three subpixels, one each for red, green and blue, and their positions are different. If there is an object on either side of the extension line of the ray, the coordinates of the intersection are calculated, and visibility is determined by z-buffering. If the intersection is visible, the color is acquired and pasted on the subpixels of the LCD. I confirmed that a simple 3D moving object, which consists of several polygons, could be rendered at more than two frames per second, and a full-parallax moving image could be obtained by using IP.

Liquid-crystal (LC) varifocal lens or LC optical deflector is useful for 3-D display. In this paper, we propose a highspeed optical deflector for 3-D display using a new LC motion. This device responds at double frequency of an applied voltage when high voltage is applied. The voltage, position and time dependences of deflection angle using square-wave or sine-wave voltage are described in this paper.

We present enhanced 3D object reconstruction of heavy occluded object and enhanced 3D object recognition using variance estimation with synthetic aperture integral imaging (SAII). SAII is a technique that acquires the elemental images moving imaging sensor of camera without the micro lens array which was used to get elemental images in conventional II system. We used volumetric II among various computational II. In volumetric reconstruction, the focused areas of the reconstructed image are varied with the distance of the reconstruction plane. A partially occluded object which is focused can be reconstructed clearly by making the occluding object blur using the lens. However, to reconstruct an unobstructed object using volumetric II in heavy occluding object, other methods are additionally required because occluded object is reconstructed with heavy blurred distribution of occluding objects. The proposed method is to get 3D unobstructed object using the value of variance of superposed pixels of elemental images for heavy occluding object. If the variance of superposed pixels is calculated, because the variance value of pixels in the only blurred distribution are small comparing with the value including the pixels of occluded object, the heavy occluded object image is reconstructed remedying blurred distribution due to the occluding objects. Calculating this variance, quality of reconstructed image is enhanced. We applied to 3D recognition with the enhanced visualization, and it is shown that the peak value of correlation with enhanced reconstruction using the proposed method is superior to the 3D recognition without enhanced 3D object reconstruction.

This paper proposes new techniques to improve image quality of the coarse integral volumetric display. Conventional volumetric displays can achieve natural 3D vision without conflict between binocular convergence and focal accommodation, while they cannot express occlusion or gloss of the objects. Multiview displays can express the latter while it cannot achieve the former. The coarse integral volumetric display can realize both natural 3D vision and expression of occlusion and gloss at the same time. Since real image or virtual image of the display panels are formed in the coarse integral volumetric display, aberration of the image can become severe. Though the author has proposed an optical setup to remove major aberration, further improvement is required to achieve better image quality. In this paper the author proposes DFD for distorted image plane, which can realize natural connections between component images. The proposed method takes into account the distortion of real/virtual image plane and each 3D pixel is drawn on the adjacent two distorted image planes so that the brightness may be in inverse proportion to the distance to each plane. Also the author discusses proper selection of component lens to improve connectivity of image.