Capturing a hologram is much easier than I expected.
I'm in Microsoft's mixed-reality capture studio in San Francisco when a director yells "Action!" Suddenly, I'm doing bad dance moves in the middle of an 8-foot circle for about 10 seconds. And that's it. With a bit of computational magic, I am now immortalized as a hologram.
There are 106 cameras around me from all angles and four microphones to capture audio. The cameras are divided equally between infrared and RGB units that capture texture, color and shape from 30 to 60 frames per second.
"We take pictures of about 10 GB / second and we carry a lot of computer vision algorithms and a lot of mesh geometric algorithms to turn it into an end product," said Steve Sullivan, executive director of the study of mixed reality capture of Microsoft.
During postproduction, the mesh can be manipulated to change the lighting or colors of the 3D model. The final product is a volumetric three-dimensional model, or hologram, compressed at 10 Mbps more manageable for reading.
When viewing the 2D model (such as the desktop screen), the viewer can change the viewpoint during execution with a mouse or a touch screen.
"We really want it to be everywhere, just like I would experience a picture or a movie," Sullivan said.
In a helmet like Microsoft HoloLens, the hologram is in your space so you can move in performance. Many people can see the hologram at the same time, but they have the right perspective depending on their position.
Currently there are three capture studios: one in San Francisco, one on Microsoft's Redmond campus in Washington and one in London.
Unfortunately, you can not enter any of these capture studios to make a hologram on your own. Microsoft responds to groups such as celebrities, advertisers, and educational service providers, such as hospitals that could use patient holograms to train doctors. Microsoft did not reveal how much a capture session would cost.
