“Our optical design could make it practical to replace 2D flat panel displays with 3D images for digital signs, enter­tainment, education and other appli­cations where 3D images provide a signi­ficant enhancement,” said research team leader Byoungho Lee from Seoul National University in Korea. “Our design could also be modified to provide immersive experiences in movie theaters, for example.” The researchers combine two different light-field display techno­logies to project large-scale 3D images with almost diffraction-limited resolution. The new display is autostereo­scopic, which means that it produces different 3D images so that the image can be viewed from various angles.

“We developed a way to carry out all the display processes optically without any digital proces­sing,” said Lee. “This compen­sates for the limi­tations of each display technology to allow the creation of high-resolution 3D images on a large screen.” Light-field displays work by repro­ducing light that is reflected from an object in a way that corresponds to the actual visible position. Because autostereo­scopic light field displays produce different images for different viewing angles, they require a huge amount of information to be processed. This demand creates a tradeoff between reso­lution and the size of the displayed image because the hardware of the display gets overwhelmed by the amount of information required.

To overcome this limitation, the researchers designed a new optical confi­guration that combines a multifocal display with integral imaging. Typically, a multifocal display can generate a high-quality volumetric image, but it is technically difficult to implement on a large-screen system. On the other hand, integral imaging is better at enlarging images. In the new design, the multi­focal display generates a high-resolution 3D, or volumetric, scene while the integral imaging techno­logy enlarges it for viewing on a large screen. The infor­mation conversion between the multifocal display and integral imaging is all performed optically without any digital processing.

“Our method goes beyond merely combining two existing methods to achieving an ultrahigh-definition volu­metric light-field display with almost diffraction-limited resolution,” said Lee. “We also found a way to effectively resolve the diffi­culty of enlarging a volumetric scene and overcame problems with infor­mation loss that tend to affect integral imaging.” After verifying the resolution of their prototype system, the researchers qualitatively confirmed that a volumetric image was reconstructed. The tests showed that the prototype can synthesize a volu­metric image of 21.4 x 21.4 x 32 centimeter, which is equi­valent to 28.6 megapixels and 36 times higher resolution than the original image.

“Our approach is very efficient at processing infor­mation, which enables a low computing cost as well as simple, high-quality, real-time system confi­guration,” said Lee. “The optical design can also be seamlessly integrated with various techniques used in existing light-field displays.” The researchers are now working to opti­mize the optics and further reduce the complexity of the multifocal display to make the projector more compact. They note that because the system is a fusion of two different techno­logies, the performance of their proposed system will likely improve as each technology develops. (Source: OSA)

Reference: Y. Jo et al.: Ultrahigh-definition volumetric light field projection, Opt. Lett. 46, 4212 (2021); DOI: 10.1364/OL.431156

Link: School of Electrical and Computer Engineering, Seoul National University, Seoul, Republic of Korea