Scientists at the Massachusetts Institute of Technology (MIT) have developed a new AI-based method to generate holograms. The technique is so efficient that it could be done on smartphones.
The science and other stuff to know
Traditionally, holograms were generated using laser beams. However, the resulting images from the technique could only be displayed as hard copies, which were difficult to reproduce. The method also couldn’t translate holograms to video — only static images.
Computer-based methods, on the other hand, can overcome these limitations but involve running physics-based simulations, which require massive amounts of data, computation, and processing power. Typically, it’ll take a supercomputer 15 to 30 minutes to generate a single hologram, and the final result might still not look photorealistic.
To speed up the process and increase the photorealistic precision of the holograms while cutting down on the computational burden, scientists at MIT have developed a new AI-based technique that can produce holograms almost instantly. The new technique, called ‘tensor holography‘, is so efficient that it could be done on a laptop or smartphone.
According to a study published in the journal Nature, MIT researchers started by creating a training dataset of 4,000 computer-generated images and their matching photorealistic holograms. Each of the images included the color and depth information for every pixel.
The team then trained tensor holography with this dataset, teaching it how to generate a hologram for every image. By the time it was done, it could generate photorealistic 3D holograms in a fraction of a second.
“People previously thought that with existing consumer-grade hardware, it was impossible to do real-time 3D holography computations,” the study’s lead author and a Ph.D. student in MIT’s Department of Electrical Engineering and Computer Science Liang Shi said in a statement. “It’s often been said that commercially available holographic displays will be around in 10 years, yet this statement has been around for decades.”
Shi and his team believe tensor holography could accelerate real-time 3D holography development timelines.
“We are amazed at how well it performs,” Wojciech Matusik, co-author of the study said.
The advancement could also fuel a spillover of holography into fields like VR and 3D printing. For instance, it could help immerse VR viewers in more realistic scenery, while eliminating eye strain and other side effects of extended VR use.
In addition, this technique could also boost the development of volumetric 3D printing. It “could prove faster and more precise than traditional layer-by-layer 3D printing, since volumetric 3D printing allows for the simultaneous projection of the entire 3D pattern,” the researchers said.
Other applications of this new technique include microscopy, medical imaging, and the design of surfaces with unique optical properties.
Tensor holography not only accelerates the process but also produced holograms with accurate occlusion and per-pixel focal control, improving images’ realism. It uses just 617 kilobytes of memory, allowing it to run interactively on low-power AI chips on smartphones and edge devices.