Some speculations about Magic Leap

.

As we all know, Magic Leap is very mysterious, take a large amount of money, but let a lot of people elusive. As Gizmodo senior reporter Sean Hollister curious intolerable, decided through various channels to try to speculate to restore the true face of Magic Leap. Take a look at his guesses and analyses:

Go straight to the subject. If I'm not mistaken, Magic Leap wants to do is an enhanced version of Google Glass, which seamlessly blends computing graphics with the real world: head-mounted devices with fiber-optic projectors, super lenses, and a very large number of cameras.

Patents and technical aspects

Professor Eric Seibel has been working on a new medical endoscope that allows the endoscope to see tiny tissues in the body. The principle of optical fiber endoscope is mainly to use piezoelectric to push the top of the fiber to rotate in a spiral pattern to scan for higher quality images. The interesting thing is that if we gather light at the top of a spinning fiber to draw images, the same system is equivalent to two optical-fiber projectors that can be used to combine other devices to create a 3D display effect.

Seibel's team is working on the technology, and they've even tried demo on some prototype helmet displays. One of the great advantages of this technique is that it's more practical and inexpensive than LCD and OLED, and it's smaller and smaller and even can be embedded in glasses.

This could be the reason Magic Leap invited Seibel's little buddy Brian Schowengerdt. Schowengerdt also helped Magic Leap get a 150,000 dollar donation from DoD to develop an ultra-high resolution optical fiber scanning display technology.

In fact, many of the patented applications of Magic leap revolve around how to reflect light from the fiber to the eye. These patents allow us to speculate on how they should move from the "reflection of the image into the eyeball" step to the "Let the sky appear a virtual entity" step.

It's hard to create an AR that's believed in the brain, and when we look at an object with our eyes, we focus on the distance between different degrees of depth, not just a flat cross-section, but the outside world reflects a lot of different angles of light coming into our eyes, which is similar to Google Glass this kind of head-wearing displays the problem of bad results: we need light from both sides and so many angles reflected into our eyes, so our perspective is relatively narrow.

In order to solve this problem, we must have equipment that can record the change of head and angle of view, so that the CG virtual objects and the light in the real world can be adjusted with the change of different angles, so as to achieve the realistic effect that the virtual object occupies a certain space.

And many of the patents of Magic Leap can solve these problems:

This patent can be used to explain how to use the LCD "occlusion mask" to block the real world in the position of virtual objects reflected in the light, so that the virtual object does not distort; This patent shows how to reflect light through the lens group Allows people to see the CG virtual objects anywhere within sight; This NASA-mandated patent explains how to use a series of device cameras to let software systems know about the position and movement of people in the real world; This patent is based on this position movement information How to adjust the display of the CG virtual object from different perspectives. That is to say, not only can we see a whale floating in the air, it should also be able to walk around the whale from all angles, and this patent describes how to use the camera embedded in the glasses to track attention. This is particularly important because Magic leap will rely on this patent to achieve the core effect: using Digital light field technology, the CG virtual object looks like it really occupies a space of depth in the real world.

Magic Leap wants us to gaze at a CG object in the virtual world like a chair in the real world. In order to achieve this effect, they seem to be based on the depth of the eyes of people to cut 3D objects into different slices, and then according to your attention points to focus on the most appropriate part of the slice to show.

Some very interesting recruiting information

I've been looking for some Magic leap job postings, and the relevant descriptions contain some interesting information: Magic Leap does not intend to make the product into a huge bulky device, in fact, they recruited a bunch of hardware engineers, ready to develop the appropriate tools to improve the existing large prototype products.

A number of recruitment messages suggest that Magic Leap may end up making a product that relies on mobile components such as smartphone styling and Android operating systems. Some of the job listings are even more specific: four nuclear mobile processors, cellular radio signals, GPS, noise cancellation headphones, 3D depth sensor cameras, and more. In fact, Magic Leap does say that this wearable device will "install a wide depth sensor" to create 3D virtual entities in the interior space-unlike Google's Project Tango.

Job Postings for voice engineers hint at some of our other Magic Leap technologies: speech recognition (voice recognition), network Truth (telepresence), field sound (live voice), and sound transfer and playback (music Transmission and reproduction). Maybe Magic leap also wants us to be able to enjoy the live performance of the real-life band in the virtual world at any time.

In addition, Magic leap may be more than just the use of Android systems, and it may take advantage of the many existing Open-source Android projects and the core that should be used to create cinematic reality interface. Magic leap may also take advantage of existing web technologies, as described in a job posting: cloud-based Applications and Services (cloud-based applications and service) to support Magic leap cinematic Reality. In another, the word "browser-based user interfaces" appeared.

Some of the Bulls in Magic Leap

In addition to the recruitment ads, I think Magic Leap existing staff also have a lot of information content. There are some very attractive big Names in this company:

1, Rony Abovitz: A guy who just put his own medical robotics company Mako Surgical to sell it for 50 million dollars.

2. Richard Taylor: A famous film special effects master, from Weta Workshop, a special effects technology company responsible for science fiction films such as Lord of the Rings, Hobbit, and Nineth district.

3, Graeme Devine: Senior game designer, former Apple game designer, Technical Group member.

4, Brian Wallace: Senior Technical Marketing Director, Samsung's "Next Big Thing" advertising campaign leader.

5, Gary bradski:open CV Creator, open CV is an open source Cross-platform Computer Vision Library, can achieve many image processing algorithms.

6, jean-yves Bouguet:google Street View team member, Indoor Street View project leader.

7, Greg Broadmore: Dr. Gebau's Science Fiction World (DR) Designer and creator of Grordbort ' s raygun-filled world.

8, Austin Grossman: "System Shock", "Deus Ex", "dishonored" and other well-known games of the author, designers.

9. John Root: Top animators with over 20 years of experience in top game and film production companies.

10. Dave Gibbons: Comic book author, artist, co-creator of the film "The Watchman" (Watchmen).

In fact, these cows have countless places to work well, but they end up with magic Leap.

So what exactly should Magic leap look like?

To synthesize the information above, Magic Leap would like to do is a lightweight other head-wearing equipment, almost the size of the glasses, there will be optical fiber connected to the projector, battery and other component components, in addition to a stack of cameras and depth sensors packaged configuration, Used to sense and identify your specific motion position information in the real world and recreate it in a virtual world. The device may run an Android system, have its own app store, and start with the gaming and interactive comic book market.

The Magic leap looked crazy. But 542 million dollars tells us that these things may really be happening. Magic Leap now may only have time to question: The product will be available only if the final product is small enough to fit.

Of course, I do not know, these are my guesses, because I have never even seen the relevant technology and products. Maybe Magic leap can't succeed, who knows. But I think that even if Magic Leap only solves a few key problems of computational graphics, the world should be spirits. I hope these things will happen.

Show me the Magic,magic Leap. Make me a believer.