publive-image

The AI technique can quickly generate holograms on smartphones using a new method called tensor holography

Intro: Massachusetts Institute of Technology (MIT) has developed a way to produce Hologram using a deep-learning method that is so efficient to run on a smartphone. The AI technique to quickly generate holograms on smartphones is called tensor holography which enables the creation of virtual reality, 3D printing, medical imaging, etc.

The AI system developed by MIT researchers finds out the best way to generate holograms from a series of input images which could soon be able to generate photorealistic 3D holograms on smartphones. A massive amount of computational power is required to generate holograms unlike the traditional VR and 3D displays, which simply produce the illusion of depth. The traditional 3D and VR technologies have major drawbacks such that they can make users feel sick. It causes nausea and eye strain because VR creates an illusion of 3D viewing although the user is staring at a fixed-distance 2D display.

Holograms could deliver an exceptional representation of the 3D world, which comprises a massive amount of data needed to generate the holograph. Holograph typically requires a massive amount of computing power. According to the MIT team, to make holographic technology more practical, they apply convolutional neural networks to the problem, creating a network capable of quickly generating holograms based on input images. Further, the team built a custom database of 4000 pairs of computer-generated images. Each pair of images, including color and depth information for each pixel, was matched with its corresponding hologram. Researchers used scenes with variable colors and shapes along with a depth of pixels to use the hologram in the new database. They used a new set of physics-based calculations to handle occlusion. Thus, Tensor holography can create holograms from images with a depth of information that can be calculated from a multi-camera setup or LiDAR sensor. The MIT team also said that the new method can create a more realistic 3D world which could probably eliminate eye strain and other effects. Besides that, visualization of medical data, microscopy, and design of 3D models are other applications of tensor holography.