Scientists from the USA have presented a fast method for generating holograms even on a regular laptop. The method is based on the work of a convolutional neural network.
Researchers have long sought to create holograms that a computer generates, but this process required a supercomputer to simulate physics, which is time consuming and produces less photorealistic results. Now researchers at MIT have developed a new way to get holograms instantly, and the deep learning method is so effective it can run on a laptop.
Scientists used to think it was impossible to do real-time 3D holography computations with existing consumer-grade hardware, said Liang Shi, lead author of the study and a graduate student in the Department of Electrical and Computer Science (EECS) at MIT.
Shi believes the new approach, which the team calls “tensor holography”, will help achieve this goal faster. This will help create holography in 3D and VR.
They used deep learning to speed up computer holography, allowing them to be generated in real time. The team designed a convolutional neural network, a processing technology that uses a chain of trained tensors to roughly mimic how a person processes visual information. Training a neural network usually requires a high quality dataset that did not previously exist for 3D holograms.
The team created a custom database of 4,000 pairs of computer generated images. To create holograms in the new database, the researchers used scenes with complex and variable shapes and colors, with uniform pixel depths from background to foreground, and a new set of physics-based calculations to handle occlusion.
Learning from each pair of images, the tensor network adjusted the parameters of its own calculations, consistently increasing its ability to create holograms. The fully optimized network performed orders of magnitude faster than physical calculations.
