To create a VR ecology on the car, Apple is developing a VR ride system

Update time: Every Monday to Saturday at 8:00 a.m. meta-universe hot news, weekly yen cosmic technology sharing.

For content cooperation & text version, please pay attention to the mia subscription account (MIA-duomeiti), or add the Shanghai Multimedia Industry Association Secretariat WeChat (mia-online).

Join the group chat - XR Future Tech World

Today's main content is:

Space situational awareness firm Slamcore completes $16 million Series A funding round

To create a VR ecosystem in the car, Apple is developing a VR ride system

Meta is building "Codec Avatars 2.0" with advanced machine learning techniques

Aiming to make a better AR headset, Google officially acquired Micro LED manufacturer Raxium

Space situational awareness firm Slamcore completes $16 million Series A funding round

British space situational awareness firm Slamcore completed a $16 million Series A round led by Robo Global Ventures and Presidio Ventures, with Samsung Ventures, Amadeus Capital, Global Brain, IP Group, MMC, Yamato Holdings, Octopus and others participating.

The market predicts that within a decade, the autonomous mobile robot (AMR) market size alone can reach $22 billion, and Slamcore believes that its AI-based positioning and mapping intelligence will play an important role in future automated navigation systems.

And with the rise of the concept of metaversals (based on VR/AR headset access), the demand for drawing and navigating both real and virtual worlds at the same time will become increasingly demanding —and it's clear that Slamcore's future will be much better.

Owen Nicholson, founder and CEO of Slamcore, said: "For a long time, robots have been unable to navigate physical spaces with the level of accuracy and efficiency we know, and as companies and consumers increasingly adopt machine intelligence technology in the coming years, Slamcore is determined to develop products based on optimized algorithms. ”

Generating accurate location perception in unmapped or rapidly changing environments, especially indoor or architectural areas with less EFFICIENT GPS, is the core problem that Slamcore set out to solve by optimizing machine mapping and moving new environments – building environmental maps in real time based solely on data collected from the machines themselves.

Although Slamcore is a software and artificial intelligence company, the company keeps a close eye on robotics and VR/AR technology and claims to have optimized its algorithms "for some of the broadest sensor and processor combinations" to enable designers and developers to use Slamcore out of the box with minimal configuration.

To create a VR ecosystem in the car, Apple is developing a VR ride system

The U.S. Patent and Trademark Office has officially granted Apple a patent related to the VR ride system, which can solve the problem of passenger motion sickness that can occur when the vehicle is driving. Designed for self-driving cars without windows, the VR headset will allow drivers to drive based on visual road views, even hold virtual meetings in the car, and enjoy in-car services such as VR games, reading, and chatting.

In order to further create a VR in-car experience, Apple has launched a variety of virtual experience scenarios designed to make passengers relax, based on the VR ride system, passengers can imagine as if they are in a boat, glider and other scenes, so as to get a unique ride experience.

The VR ride-hailing system enables passengers to initiate virtual experiences based on real physical location or virtual city or scenery, and it is worth mentioning that these virtual experiences are both educational and interactive, for example, key information such as historical monuments or landmarks can be marked and annotated during the passenger visual experience of the city map, not only that, in the zombie apocalypse scene, the user encounters traffic lights when driving away, and can also simulate interactive scenarios such as real traffic road conditions.

In some examples, the VR projection device 720 and / or VR controller 710 may include one or more microphones for receiving voice input from the passenger; e.g., the headset is built-in to receive a microphone from the main control passenger voice input, the voice input may be used for VR system voice control, or for communication with other passengers, or for telephone calls and conference calls through the VR controller or ride system such as external communication.

Further, in other examples, the VR projection device and / or VR controller may include or coupled to an internal camera or other sensor for detecting the passenger's hand, arm or head posture, eye movements, facial expressions and other postures, some examples show that the detected pose, action or expression may be used as an input mode of the VR system to affect the presentation and / or display of virtual content to individual passengers.

Figure 1 of apple patent shows an example of a driver wearing a VR headset to operate a VR ride system

Apple patent Figure 2 shows an example of a VR ride system that projects VR content into a vehicle window for passengers to see

Meta is building "Codec Avatars 2.0" with advanced machine learning techniques

Meta researchers are building "Codec Avatars 2.0" through advanced machine learning techniques.

Facebook (Meta) first demonstrated "Codec Avatars" in March 2019, which is powered by multiple neural networks, generated based on 171 camera captures, after which "Codec Avatars" will be driven in real time by a prototype VR headset configured with 5 cameras (2 cameras for internal observation and 3 cameras for external observation), and as the version iterates, The researchers have demonstrated systems such as more realistic eye tracking and microphone-enabled demos.

In April, at the MIT "Virtual People and Virtual Beings" workshop, Yaser Sheikh, team leader of Codec Avatars, showed off the project's latest demo, Codec Avatars 2.0.

Sheikh said: "I think one of the big challenges in the next decade is, can we achieve immersive remote interaction? ”

In a paper published last year, Sheikh's team claimed that Codec Avatars' latest model will be smaller and more efficient, and that the neural network will now only compute pixels visible to the headset, and with this advance, the Quest 2 headset will be able to render five avatars in an empty scene at 50 frames per second.

However, to get to the level of fidelity, the Meta R & D team still has some way to go, today's avatar presents a cartoon image, very suitable for large VR social application environments such as "Horizon Worlds".

In an interview with Lex Fridman, Zuckerberg described a future in which users could use "expressionist" avatars in casual games and "realism" avatars in work sessions.

It is worth mentioning that during the workshop, Sheikh pointed out that it is impossible to predict the birth of the final form of Codec Avatars, but Sheikh believes that Codec Avatars is the "miracle light" of the company's project.

Aiming to make a better AR headset, Google officially acquired Micro LED manufacturer Raxium

Google confirmed that the company has formally acquired Raxium, a 5-year-old startup known for its MicroLED technology, and it is clear that Google's acquisition of Raxium is dedicated to making the next generation of VR/AR/MR headsets.

It's further evidence that Google will make a big push into AR after Google acquired eyewear maker North in 2020; Google is reportedly hiring engineers to build an AR ecosystem operating system, and in January, Google Labs was accused of building an AR headset called "Project Iris" that operated in the same way as the "Project Starline" high-resolution video chat demo shown at the I/O event last year.

Last month, The Information first reported Google's acquisition of Raxium, noting that MicroLED technology could help build AR displays that are more energy efficient and rich in color than other solutions, and that Raxium is working on MicroLED "monolithic integration" technology, The Information said, which means that batch processors made of the same type of silicon material will likely lower prices.

In addition to Raxium, companies including Apple and Vuzix are also working on microLED AR hardware research, the AR market, Google AR products will accept strong competition from Microsoft HoloLens, while Apple, Meta, Snap and other companies are also investing heavily in self-developed immersive hardware that can present images and information in the real world, and in the future, the competition in this track will become more and more intense.

Raxium said that the pixel pitch (the distance between the center of one pixel and the center of another pixel) is about 50 microns, while the pixel pitch of its MicroLED screen is about 3.5 microns, and it has "unprecedented display efficiency" that is more than five times faster than the current record.

93913 Feed

END