Volumetric 3D display provides true perception of objects
The physical world around us is in 3D, yet traditional display devices are only able to show 2D flat images that lack the depth (third dimension) information. This fundamental restriction greatly limits the ability of human beings to perceive and understand the complexity of real-world objects. Nearly 50% of the human brain's capability is devoted to processing visual information. Flat images and 2D displays do not harness the brain's power effectively.
There are four major physical depth cues the human brain uses to gain true 3D sensation. The first, accommodation, measures how much the eye muscles force the eye lenses to change shape when focused on an image of a specific 3D object in a scene. The second, convergence, measures the distance eyes have to cross to see a 3D object simultaneously. The third, motion parallax, offers depth cues by comparing the relative motion of different elements in a 3D scene. And finally, binocular disparity refers to differences in images acquired by the left and right eyes.
Traditional 2D displays fail to provide these 3D depth cues to viewers, and that often leads to ambiguity and confusion in high-dimensional data/graphics presentations. By contrast, volumetric 3D display technologies can provide all the above-mentioned depth cues by displaying 3D volumetric images in a true 3D space. Each voxel on a 3D image (analogous to a pixel in 2D image) locates physically at the spatial position where it is supposed to be, and emits light from that position in all directions to form a real 3D image in 3D space. Volumetric 3D display technologies yield realistic spatial representations of 3D objects and simplify our understanding of complex real-world 3D objects and the spatial relationships among them.
Figure 1 illustrates the principle of our digital-light-processing (DLP)/Helix volumetric 3D display.1–4 Light coming out from a source is reflected by a polarizing beamsplitter cube towards a spatial light modulator (SLM), whose image patterns are generated by a host personal computer (PC). The modulated image patterns are projected by optics onto a rotating double helix screen. We use a DLP chip set provided by Texas Instruments (TI) as the high-speed SLM device in our system. The DLP projection engine provides high-speed, high-resolution, and high-brightness image generation for volumetric 3D displays.
As part of the image generation process, we synchronized the rotating helix screen's motion with the DLP pattern projection's timing such that the moving screen intercepts high-speed 2D image projections from the DLP at different spatial positions along the z-axis. This forms a stack of spatial image layers that viewers can perceive as true 3D volumetric images. Viewing such images requires no special eyewear. The 3D image is floating in true 3D space, just as the real object is placed there. The unique features of the DLP/Helix 3D display design include an inherent parallel architecture for voxel-addressing, high-speed and high-spatial resolution, and no need for viewers to wear special glasses or a helmet.
A key component in the DLP/Helix 3D display system is an interface board between the host PC and DLP chip. We designed and built a high-speed DLP image projection engine (dubbed the H3D): see Figure 2. It is able to directly connect the digital visual interface port on the host PC with the DLP chip, providing dynamic 3D image display capability and a streamlined configuration.
A system prototype can display volumetric 3D images with a resolution of more than 150 million voxels. Figure 3 shows a photo of a sample 3D image displayed on the DLP/Helix system. Unfortunately, since the photo is 2D, the real volumetric 3D effect is lost in the 2D paper print. The system is also able to display dynamic 3D image sequences such as 3D video. The 3D images displayed on our volumetric 3D monitor have a see-through transparency feature that lets viewers see both the foreground and background structures, thus increasing the efficiency of understanding 3D spatial relationships.
In conclusion, we designed and prototyped a volumetric 3D display based on a DLP projection engine, with image resolution topping 150 million voxels. The volumetric 3D display is fundamentally different from the conventional 3D rendering visualization technique where the object is displayed on a 2D flat screen with 3D rendering for depth perception. It is also different from 3D stereo video or a head-mounted display where the 3D perception is created with a pair of polarized or switchable glasses or display screens. A volumetric 3D display device projects 3D images directly into a true 3D volume in space that can be viewed by the naked eye without the need for any special 3D glasses. Viewers can walk around a 3D image, look at it from different angles, and get a realistic sense of depth, just as though they were looking at the real physical object. Such a 3D display provides both physiological and psychological depth cues to human viewers for truthfully perceiving objects in 3D space.
The current prototype system displays monochromic 3D images. In the future, we will develop a full-color version of the display system with a sufficient 3D image refresh rate. We will also work with potential users to develop effective 3D user interface techniques for the system.
Jason Geng has 20 years of experience developing 3D imaging technologies. He has published more than 100 papers and one book. He is an inventor who holds 23 issued patents. He has received prestigious national honors, including the Tibbetts Award, and was named to the Inc. 500 list of top entrepreneurs. He currently serves as vice president for the IEEE ITS Society.
