Open Access
Add to BibSonomyAbstract
Lamina 3D display is a new type of multi-layer 3D display, which utilizes the polarization state as a new dimension of depth information. Lamina 3D display system has advanced properties – to reduce the data amount representing 3D image, to be easily made using the conventional projectors, and to have a potential being applied to the many applications. However, the system might have some limitations in depth range and viewing angle due to the properties of the expressive volume components. In this paper, we propose the volume using the layers of switchable diffusers to implement the active-type Lamina 3D display system. Because the diffusing rate of the layers has no relation with the polarization state, the polarizer wheel is applied to the proposed system in purpose of making the sectioned image synchronized with the diffusing layer at the designated location. The imaging volume of the proposed system consists of five layers of polymer dispersed liquid crystal and the total size of the implemented volume is 24x18x12 mm3. The proposed system can achieve the improvements of viewing qualities such as enhanced depth expression and widened viewing angle.
© 2015 Optical Society of America
1. Introduction
Recently, three-dimensional (3D) display become a hot topic in both research and industry fields. The studies about the 3D display span in various directions, and it is the critical issue to overcome the disadvantages of conventional 3D display methods such as the 3D nausea caused by accommodation-vergence mismatch and the uncomfortableness by wearing the additional equipment. The main goal of 3D display is to provide the sense of reality to viewers with the whole 3D depth cues of the solid objects such as binocular disparity, vergence, accommodation, and motion parallax. The stereoscopic 3D display, one of the most conventional 3D display methods at present, represents 3D image using only the binocular disparity, which causes the disagreements with the other 3D perception cues. The best way for completely representing 3D images which induce all perception cues is holography. However, as is well known, the required quantity of information is too huge to represent realistic 3D images using holographic display systems [1–3]. Hence many researchers pay attention to the volumetric 3D displays which can represent more natural 3D images than conventional stereoscopic 3D displays and require less amount of processing data than holographic displays [4, 5].
The volumetric displays can present 3D images in the specific physical space. A broad range of depth cues is inherently associated with such images, including the parallax and the oculomotor cues [6]. The volumetric 3D images are expressed by voxels, i.e. volume pixels [7, 8]. The implementation of display volume can be accomplished using the mechanical instruments, where the rotational or the translating movement sweeps through the volume under the afterimage effect. The multi-layer 3D display is one of the volumetric 3D displays and composed of the multiple layers which display the layered images sectioning the 3D volume. The typical multi-layer 3D displays are depth-fused display (DFD) and Depth Cube [9–11]. The DFD is one of the simplest multi-layer 3D displays, which consists of two layers of 2D displays. The DFD system represents 3D images between two displays by adjusting the intensity ratio between the layers [9]. The DFD has a disadvantage about overlapped arrangement causing the limitation of the viewing angle. Moreover, the transparent displays or beam splitter must be used to compose the layers in the DFD system, where the transparent display has a high cost and is not practical in this time, and beam splitter needs a large space for the whole display system. For solving this problem, Depth Cube is proposed. Depth Cube is a progressive multi-layer 3D display system, which is composed of twenty multiplanar optical elements and a high-speed digital light processing (DLP) projector. Depth Cube generates 20 sliced images within the time giving the afterimage effect using deformable mirror device and computer graphic techniques. However, since the refresh speed and the displayed color are closely related in DLP, the color depth of 3D images reconstructed by Depth Cube is severely degraded. Therefore, new multi-layer 3D display system is demanded to resolve a previous disadvantages which are restricting to the layers, requiring the additional high technology equipment, and using the additional process and relatively huge amount of information.
There are some other volumetric display methods with deformable mirror device (DMD) and special volumes using the high speed performance of DMD projector. The special volumes can be implemented by using a rotating diffusing screen or streaming of the water drops [12, 13]. Moreover, multi-layer 3D displays based on the reconstruction of light field rather than depth-fusing, named tensor displays, are also proposed [14–16]. With the compressive optimization technique of the light field passing the multiple attenuation layers, the tensor display can reconstruct the directional views based on light field. In the tensor display system, the image processing is complicated and requires high computational performance such as graphic processing unit acceleration for the demonstration of video images.
In the previous papers, we proposed new type of multi-layer 3D display system, named Lamina 3D display, which adopts the polarization ratio to process the depth information [17, 18]. The proposed method using the modulated polarization by the spatial light modulator (SLM) changes the rotation ratio of polarization of each pixel according to the gray scale, which is decided by the applied voltage of the pixel. The Lamina 3D display system can represent more natural 3D information using less complex system structure compared to the conventional volumetric 3D display systems. The volume of Lamina display consists of the multiple layers of scattering polarizer films (Imajor, Teijin DuPont, Japan), which scatter and transmit the incident light according to the angle difference between the polarization axes of the light and the film. However, the polarization dependent scattering property of the film produces blurring in the reconstructed images. Because the polarization of light is separated into the only two orthogonal components, the polarization states cannot be divided in the intermediate conditions except the orthogonal case [18, 19]. This orthogonal property of polarization brings the results of polydispersion between layers, and it causes a limitation of increasing the number of layers and blur in the 3D reconstructed images, which results in the degradation of resolution. If the number of scattering polarizers is increased, the image blur can be worse and induce further limitation on the depth range and viewing angle of 3D images. The reconstructed 3D images of the Lamina display express the volumetric characteristics within the laminated volume, so the additional layers are required to expand the volume. This contradiction of number of layers is an innate problem and limitation of polarization decoding, which must be relieved and resolved to improve the overall viewing qualities.
In this paper, we propose a new decoding method of Lamina display system using active screens, named active-type Lamina display, to improve the viewing property by increasing the number of layers without blurring. The active-type Lamina system is proposed to increase the number of volume layers without any change of the encoding part of the passive-type Lamina display. Compared to the previous Lamina display system, the decoding part of the proposed system is composed of a rotating polarizer wheel and polymer dispersed liquid crystal (PDLC) screens instead of the scattering polarization films [20]. The proposed system can represent the less blurry 3D images because PDLC screens are turned on in sequence according to the designated image sectioned by the polarizer wheel, and it can reduce the interference between the layers. Although the time-multiplexing scheme is adopted to the decoding part, the proposed system does not suffer from the conventional disadvantages of time-multiplexed 3D displays because the encoding and the decoding parts are separated in the operation. We prove the feasibility of the proposed system through the implemented experimental setup, which can present enhanced depth expression and widened viewing angle.
2. Active-type Lamina 3D display system
2.1 Configuration of the proposed system
The principle and configuration of Lamina display system are illustrated in Fig. 1 . As mentioned in the previous chapter, the encoding parts of the passive-type and the active-type Lamina display are just the same, which are composed of projection-type display system for 2D image, linear polarizer, SLM for modulation of polarization, and projection optics. The encoding part of the system plays a role of converting the depth information into the polarization status based on polarization distributed depth map (PDDM) technique.
The first stage of the encoding process is the alignment of a 2D image displayed from the projection-type display as an initial polarization condition of 2D image by passing through the linear polarizer. The direction of alignment is associated with the initial state of the SLM which is the Twisted Nematic (TN) liquid crystal (LC) panel without polarizers and can modulate the polarization status pixel by pixel according to the given depth information. The projected 2D image must be focused on the SLM, whose size is also adjusted to be matched to the size of SLM. The depth information is provided to TN-LC SLM as a grayscale depth map. In general, the TN-LC panel represents the grayscale through the rotating modulation of the polarization status of incident back light according to the applied voltage. Since the polarizers on the SLM are removed, the polarization status is modulated according to the depth map in the Lamina encoding system instead of the intensity of the incident light, which is the 2D image projected to the SLM. Therefore, the 2D image of which the polarization status is rotated pixel by pixel according to the depth map can be obtained as the output of the encoding part of Lamina display system.
In the passive-type Lamina system, the decoding process is performed by the multiple layers of the scattering polarizers, which induce the severe blurring. In the proposed system, PDLC layers are substituted for the scattering polarizers to represent the image at the designated location. Because PDLC layers are not influenced by the polarization status, the additional depth sectioning process is necessary to decode the PDDM information. The PDDM image can be easily converted into sectioned 2D images by passing a linear polarizer whose angle determines the depth position of sectioning. The rotating polarizer wheel is used in the proposed system, and it is segmented into polarizer windows and blocking masks as shown in Fig. 2(a) . The polarization angle of each window is adjusted to match the location of corresponding PDLC. The masks block the intermediate images that can be the unintended interference during the switching time of PDLC. The markers of the wheel are set up for the synchronization between the wheel and the PDLC layers to place the sectioned 2D images on the designated positions. Figure 2(b) illustrates the configuration of the synchronization process.
Fig. 2 (a) Rotating polarizer wheel and (b) synchronization process between the polarizer wheel and PDLC.
The PDLC panels used for the volume layers consist of mixture of ultraviolet (UV) light-sensitive polymer (NOA89, Norland Products Inc.) and liquid crystal (ZKC 5095, Chisso). The mixture is injected between indium tin oxide (ITO) coated glasses, and cured with UV light. Table 1 shows the transparency and response time of the PDLC layers according to the applied voltage. In the “field-on” state, the layers become transparent when the voltage is applied. The PDLC layer in the “field-off” state can be used as the diffuser screen.
Table 1. Transmittance and response time of PDLC layers
2.2 Simulation of depth expression
In the active-type Lamina system, a volumetric 3D image is synthesized from the sectioned 2D images of each layer which are represented in sequence. Considering the temporal resolution of the human eye, the frequency for one sweeping cycle of total volume layers should be more than 24 frames per second (fps), which determines the proper number of volume layers with the response time of PDLC. In the proposed system, the expressible volume is composed of five layers, which needs about 8 ms for the operation time of each layer, and it agrees with the characteristics of the given PDLC layer.
The image distributed in the volume layers can be observed as a certain depth image by virtue of the depth-fusing effect, which determines the depth location by intensity distribution over layers. Figure 3 shows five steps of depth map and the sectioned 2D images obtained from the PDDM image by passing through the linear polarizer of the specific angles. As shown in Fig. 3, the dark strip seems to shift through all steps during one polarization angle period from 0° to 90°.
Fig. 3 (a) 5 steps depth map, and sectioned 2D images from PDDM image obtained by polarization angle as (b) 0°, (c) 23°, (d) 45°, (e) 68°, and (f) 90°.
The actual appearance of 3D image represented by the proposed system can be predicted using the sectioned images by Monte Carlo simulation. Based on the polarization sectioning, we calculated the intensity distribution of each layer, and we image the random ray bundle by a lens so that we can find the beam waist whose location is considered as an imaging point of depth-fused image [21]. Since the total overlapped intensities due to the depth position could not be same, the observed intensity level for each step could be different. Figure 4 shows the comparisons between the simulated results and the actual views of the stepped depth maps, which are observed at the left upper side view to recognize the depth. As shown in Fig. 4, the simulation describes the depth and the intensity representation of the actual system exactly. The simulation results can be applied to the compensation for the 3D image of the proposed system.
Fig. 4 Comparison between simulations and actual system views. (a), (b), and (c) present the simulation results for 3, 4 and 5 steps depth maps respectively, while (d), (e), and (f) show the reconstructed images of the proposed system for the similar conditions of simulations.
3. System setup and results
To implement the active-type Lamina display system, liquid crystal on silicon (LCoS) projector (XEED 600X, Canon, Japan) of which the projection optics is removed is used as the 2D image projection system. The depth-encoding SLM is extracted from LCD projector (PT-LM1E-C, Panasonic, Japan). For the relay optics, we use two Nikon 50 mm 1:1.4D lenses. The polarization wheel has ten polarizer windows, of which the polarization angles are 0°, 23°, 45°, 68°, and 90°, and the angles are repeated for the other half. The inner radius of the polarizer wheel is 12 mm, and the outer radius is 18 mm. The driving motor (PM25L-024, Minebea, Japan) can rotate the polarization wheel at 83.33 ms for one turn, which makes 24 times sweeping for one second the five layers of the expressible volume synchronized with the wheel. Therefore, the proposed system can represent the reconstructed 3D image at 24 fps within the boundary of afterimage effect. Figure 5 illustrates the configuration and setup of the active-type Lamina display system while Table 2 shows the summarized specifications of the system implementation. We should remark that though the PDLC can be turned on and transparent about 100 V, we apply 150 V square waves at 1 kHz to obtain sequential driving of the five PDLC screens for faster response and higher transmittance in the actual operation. Besides, when the field-on and field-off states of PDLC are set to 95% and 30% of transmittance on each state, field-on response time (τ on) and field-off response time (τ off) can be reduced to each about 1.2 ms and 2 ms. Actually, when the cycle of the driving signals is less than 3 ms, the change of the states is not observed. The number of PDLC layer can be increased as the response time of PDLC panel is reduced. In other words, the response time of used PDLC panels allows only 5 layers for the implemented imaging volume. In Depth Cube, 20 layers of PDLC panels of which response time is sub millisecond order form the multi-layered imaging volume [10]. As mentioned above, in the proposed system, the depth information is decoded using the polarization status which is based on only two independent axes. Therefore, there is the inherent limitation in the resolution of depth expression in the Lamina 3D display, and the excessive number of PDLC layers cannot improve the viewing quality of the reconstructed image. If the switching performance of the PDLC panel can be enhanced in the response time and the transparency, the proper number of PDLC layers can be increased in the future studies.
Table 2. Transmittance and response time of the PDLC layers
As shown in Table 2, the total size of the implemented imaging volume is 24 × 18 × 12 mm3. The volume can be expanded according to the size of PDLC panel. For example, 20-inch PDLC panels are used for Depth Cube [10], which can be directly applied to the proposed system. The footprint of the proposed system is determined by the size of 2D image system and projection-type relay optics. Therefore, the space for the proposed setup can be reduced and become efficient to adopt a small-size LCoS projector as the 2D image system. As shown in Fig. 5, the proposed system can be assembled using the conventional projector and the simple optical and electrical components without expensive instruments. Therefore, the proposed system can be implemented cost-effectively.
Figure 6 (including movies) shows the results of the proposed system operation. As shown in Fig. 6, the stripe images used for depth calibration, the simple geometric images, and the car image are demonstrated with the depth map images and the various viewpoints. The results show the parallaxes of the represented images due to the continuous viewpoints and prove the feasibility of the proposed method as a volumetric 3D display. There are some color distortions and luminance deviations due to the SLM for depth decoding as shown in Fig. 6. These problems are caused by the characteristics of LC SLM such as the nonlinear response curve and the partial circular polarization rotation. The calibration of the color coordinate of 2D image system can be a solution for the color distortions. The modified polarizer wheel of which angles of polarizers are adjusted to the optimum directions can be adopted to alleviate the luminance deviations.
Fig. 6 Reconstructions of active-type Lamina 3D display. (a), (e), and (i) show the depth map, while (b) – (d), (f) – (h), and (j) – (l) show the images from the left, the center, and the right views of the reconstructed 3D images respectively. Media 1, Media 2 and Media 3 show the parallax changes of results of upper, middle, and lower images, (c), (g) and (k) respectively.
4. Conclusion
We propose the active-type Lamina 3D display of which the expressible volume is composed of five PDLC layers for alleviating the blurring problem of the previous passive-type system. The number of volume layers can be increased in the active-type system, which is influenced by the response time of PDLC panel. The wider viewing angle and the deeper depth expression need higher speed and better transition rate between transmittance and scattering of the PDLC panels. The number of layers in our implementation is determined by the specification of the PDLC, which also decides the other specifications such as the polarization wheel and the rotating motor.
The merits of the proposed system can be shared with the passive-type Lamina 3D display system, which is relatively simple and can be achieved using low data amount and few apparatuses compared with other volumetric 3D systems. Also high compatibility of depth map based 3D data can be advantageous for preparation of 3D contents [22, 23]. Therefore, we believe that the proposed method can be an excellent candidate for a next-generation 3D display.
Acknowledgments
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2011-0030079).
The computer graphic car object shown in Fig. 6 and Media 3 is based on the model made by natman and is used under Creative Commons Attribution 3.0. The human figure shown in Figs. 5 is based on the model created by Lunarmoonable and is used under Creative Commons Attribution 3.0.
References and links
1. T. Okoshi, “Three-dimensional displays,” Proc. IEEE 68(5), 548–564 (1980). [CrossRef]
2. E. Downing, L. Hesselink, J. Ralston, and R. Macfarlane, “A three-color, solid-state, three dimensional display,” Science 273(5279), 1185–1189 (1996). [CrossRef]
3. E. B. Goldstein, Sensation and perception (Wadsworth 2007).
4. R. Hirayama, M. Naruse, H. Nakayama, N. Tate, A. Shiraki, T. Kakue, T. Shimobaba, M. Ohtsu, and T. Ito, “Design, implementation and characterization of a quantum-dot-based volumetric display,” Sci. Rep. 5, 8472 (2015). [CrossRef] [PubMed]
5. W. Song, Q. Zhu, T. Huang, Y. Liu, and Y. Wang, “Volumetric display based on multiple mini-projectors and a rotating screen,” Opt. Eng. 54(1), 013103 (2015). [CrossRef]
6. J. Chen, W. Cranton, and M. Fihn, Handbook of Visual Display Technology (Springer 2012).
7. B. Blundell and A. Schwarz, Volumetric Three-dimensional Display Systems (Wiley 2000).
8. H. H. Refai, “Static volumetric three-dimensional display,” J. Disp. Tech. 5(10), 391–397 (2009). [CrossRef]
9. S. Suyama, S. Ohtsuka, H. Takada, K. Uehira, and S. Sakai, “Apparent 3-D image perceived from luminance-modulated two 2-D images displayed at different depths,” Vision Res. 44(8), 785–793 (2004). [CrossRef] [PubMed]
10. A. Sullivan, “A DepthCube solid-state 3D volumetric display,” Proc. SPIE 5291, 279–284 (2004).
11. M. Date, T. Hisaki, H. Takada, S. Suyama, and K. Nakazawa, “Luminance addition of a stack of multidomain liquid-crystal displays and capability for depth-fused three-dimensional display application,” Appl. Opt. 44(6), 898–905 (2005). [CrossRef] [PubMed]
12. A. Jones, I. McDowall, H. Yamada, M. Bolas, and P. Debevec, “Rendering for an interactive 360° light field display,” ACM Trans. Graph. 26(3), 40 (2007). [CrossRef]
13. P. Barnum, S. Narasimhan, and T. Kanade, “A multi-layered display with water drops,” ACM Trans. Graph. 29(4), 76 (2010). [CrossRef]
14. D. Lanman, G. Wetzstein, M. Hirsch, W. Heidrich, and R. Raskar, “Polarization fields: dynamic light field display using multi-layer LCDs,” ACM Trans. Graph. 30(6), 186 (2011). [CrossRef]
15. G. Wetzstein, D. Lanman, W. Heidrich, and R. Raskar, “Layered 3D: tomographic image synthesis for attenuation-based light field and high dynamic range displays,” ACM Trans. Graph. 30(4), 95 (2011). [CrossRef]
16. H. Gotoda, “Implementation and analysis of an autostereoscopic display using multiple liquid crystal layers,” Proc. SPIE 8288, 82880C (2012).
17. S. G. Park, S. Yoon, J. Yeom, H. Baek, S.-W. Min, and B. Lee, “Lamina 3D display: projection-type depth-fused display using polarization-encoded depth information,” Opt. Express 22(21), 26162–26172 (2014). [CrossRef] [PubMed]
18. S. G. Park, J.-H. Kim, and S.-W. Min, “Polarization distributed depth map for depth-fused three-dimensional display,” Opt. Express 19(5), 4316–4323 (2011). [CrossRef] [PubMed]
19. P. Yeh and C. Gu, Optics of Liquid Crystal Displays (Wiley, 2010)
20. J. W. Doane and G. P. Crawford, Polymer Dispersed Liquid Crystals (Wiley, 1990).
21. S. Ravikumar, K. Akeley, and M. S. Banks, “Creating effective focus cues in multi-plane 3D displays,” Opt. Express 19(21), 20940–20952 (2011). [CrossRef] [PubMed]
22. M. Cho and D. Shin, “Depth resolution analysis of axially distributed stereo camera systems under fixed constrained resources,” J. Opt. Soc. Korea 17(6), 500–505 (2013). [CrossRef]
23. G. Li, K.-C. Kwon, G.-H. Shin, J.-S. Jeong, K.-H. Yoo, and N. Kim, “Simplified integral imaging pickup method for real objects using a depth camera,” J. Opt. Soc. Korea 16(4), 381–385 (2012). [CrossRef]
