Nov 29

Chinese regulators have announced new rules governing video and audio content online, including a ban on the publishing and distribution of “fake news” created with technologies such as artificial intelligence and virtual reality. From a report: Any use of AI or virtual reality also needs to be clearly marked in a prominent manner and failure to follow the rules could be considered a criminal offense, the Cyberspace Administration of China (CAC) said on its website. The rules, effective Jan. 1, were published publicly on its website on Friday after being issued to online video and audio service providers last week. In particular, the CAC highlighted potential problems caused by deepfake technology, which uses AI to create hyper-realistic videos where a person appears to say or do something they did not. Deepfake technology could “endanger national security, disrupt social stability, disrupt social order and infringe upon the legitimate rights and interests of others,” according to a transcript of a press briefing published on the CAC’s website.

Read more of this story at Slashdot.

full article

Nov 26

In a Quartz article, Adam Epstein writes about the filmmaking technology used to film The Mandalorian on Disney+: Industrial Light & Magic (ILM) — the Lucasfilm subsidiary George Lucas founded in 1975 to make the visual effects for Star Wars — deployed a real-time 3D projection system called “Stagecraft” on the Disney+ series that could, eventually, replace green-screen as the film industry standard for rendering virtual environments. The company has been testing Stagecraft for five years — most recently on the Star Wars spin-off movie Solo in 2018. But The Mandalorian, the flagship series on Disney’s new streaming service, likely marks the most extensive use yet of the new system.

Stagecraft’s chief innovation is that it can project a 3D visual environment around the actors that changes in real time to match the perspective of the camera. When the camera moves, the background moves too, simulating the experience of filming in a different location. It’s a significant upgrade from green-screen technology, which requires the filmmakers layer in a static image or footage after filming in front of the blank backdrop. […] The tech has a wide range of benefits. For starters, it can draw better performances from the actors, who don’t have to imagine the environment they are in, as they do when filming in front of green-screen. They can instantly be transported to any location, real or made-up, and feel as though they are there. And that’s another big advantage: Stagecraft allows films and TV shows to simulate environments without actually having to send an entire production there to film. “One downside is that the displays used in Stagecraft require liquid crystals that take several years to grow,” the report adds. “Growing and maintaining these crystals, which are the backbone of LCD (liquid crystal display) screens, can be expensive and time-consuming, perhaps complicating the attempts of other companies to adapt the technology.”

This video from Unreal Engine shows a smaller scale version of the tech in action.

Read more of this story at Slashdot.

full article

Nov 25

Amazon’s annual AWS re:Invent conference in Las Vegas — where the tech giant reliably announces a host of products heading to Amazon Web Services, its cloud platform — doesn’t kick off officially until next week. But that didn’t stop the tech giant from previewing a few of the highlights, the bulk of which relate to the internet of things (IoT). From a report: Why the investment in IoT? Perhaps because AWS maintains pole position in the segment, which is anticipated to be worth $212 billion by the end of 2019. Amazon CTO Werner Vogels told VentureBeat in a recent interview that AWS customers deploy upwards of hundreds of thousands of sensors. First on the list was Alexa Voice Service (AVS) Integration for AWS IoT Core, the managed cloud service that lets gadgets interact with cloud apps and other devices. It’s designed to let manufacturers create Alexa built-in devices — or accessories that connect to Alexa to play music, control smart home devices, and more — with constrained hardware resources. Alexa built-in devices previously required at least 100MB of RAM and ARM Cortex A-class microprocessors, but thanks to new AWS cloud processing components that offload tasks like buffering and mixing audio, the baseline requirement has been reduced to 1MB of RAM and Arm Cortex M-class microcontrollers. Alexa Voice Service (AVS) Integration for AWS IoT Core specifically offloads media retrieval, audio decoding, audio mixing, and state management to a new virtual Alexa-built in device in the cloud. New AWS IoT-reserved MQTT topics allow for message transfer between devices connected to AWS IoT Core and AVS using the MQTT protocol.

Read more of this story at Slashdot.

full article

Nov 24

An anonymous reader quotes the International Business Times:

At the recent Tianfu cup held in Chengdu, China, Chinese China’s top white-hat hackers have converged to test zero-days against top software available in the market today. During the first day of the event, Chinese security researchers were able to break into major browsers such as Safari, Microsoft Edge, and Google Chrome.

Since March 2018, the Chinese government has officially discouraged security researchers from joining hacking competitions outside the county. The recent Tianfu Cup is the venue for hackers to showcase their skills and even earn six-figure bounties for successful exploits. Former Pwn2Own winner Team 360 Vulcan took home $382,500 for successfully hacking the old version of Office 365, Microsoft Edge, Adobe PDF Reader, VMWare Workstation, and gemu+ Ubuntu during the two days event, reports ZDNet… Search engine giant Google has a representative in the event with some members of the Google Chrome security team present on site. Organizers plan to submit a report of all bugs uncovered during the event to all vendors when the competition concludes, says ZDNet.

Read more of this story at Slashdot.

full article

Nov 21

An anonymous reader quotes a report from Ars Technica: After a tease earlier this week, Valve has revealed more details and a new trailer for the first new Half-Life content in over a decade. The “full-length” Half Life: Alyx will hit Steam in March 2020, Valve says, with support for “all PC-based VR headsets.” Pre-orders are already available for $59.99, though the game will be free if you own a Valve Index headset. The game, which Valve says is “set between the events of Half-Life and Half-Life 2,” has been “designed from the ground up for Virtual Reality” (i.e. you can stop hoping for a 2D monitor release). “Everyone at Valve is excited to be returning to the world of Half-Life,” Valve founder Gabe Newell said in a statement. “VR has energized us.”

Today’s video trailer shows that next year’s Alyx-ization of Half-Life is equal parts abstract and concrete. The VR perspective from today’s trailer doesn’t include any floating body parts or feet; the only part of your virtual self you’ll see, at least in today’s trailer, is your hands, covered in a pair of gloves. Yet we also hear Alyx’s voice, which indicates that this game’s protagonist won’t be nearly as silent as Freeman in his own mainline adventures. Today’s announcement includes video footage that confirms a data-leak examination by Valve News Network earlier this year: a new manipulation system dubbed the Gravity Gloves. And boy do these things look cool. Need to grab or pick something up? Point at whatever that object is (whether it’s close or far away) with an open hand until it glows orange, then close your hand and flick your wrist toward yourself to fling the item in your direction. At this point, you get a moment to physically “catch” the object in question. Point, clench, flick, catch.

Today’s trailer also confirms bits and pieces of the exciting HLA details I’ve previously heard about from multiple sources. For instance, the trailer includes teases of the game’s approach to VR-exclusive puzzles, particularly those that require moving hands around a three-dimensional space. Some of these puzzles will require scanning and finding clues hidden inside of the virtual world’s walls (and moving or knocking down anything hindering your ability to see or touch said walls). Other puzzles will require arranging what look like constellations or grids of stars around a 3D space in order to match certain patterns. And then there’s the matter of familiar Half-Life creatures coming to life for the first time in over 12 years, which means they’re that much more detailed and gruesome as rendered in the Source 2 engine. The Half-Life website specifies that this game can be played sitting, standing, or with “roomscale” movement. Players can use finger-tracking or trigger-based VR controllers and move around the VR environments by “teleporting” from point A to point B, “shifting” smoothly to a new position, or just walking continuously with an analog stick.

Read more of this story at Slashdot.

full article

Nov 20

Yesterday, Valve announced Half-Life: Alyx, the first new game in the acclaimed Half-Life series in well over a decade. And unlike the previous Half-Life installments, this game will be playable exclusively in virtual reality. The Verge reports: We don’t currently have any details beyond the tweet from Valve above, which appears to be the first tweet from a new, Twitter-verified Valve Software account established in June. But clearly, we’ll be learning more on Thursday, presumably from this social media account, at 10am PT. Despite being some of the most influential and critically acclaimed PC games ever made, Valve has famously never finished either of its Half-Life supposed trilogies of games. After Half-Life and Half-Life 2, the company created Half-Life: Episode 1 and Half-Life: Episode 2, but no third game in the series. The closest we’ve come to knowing anything about where Half-Life was headed was this thinly veiled fanfic from former Valve writer Marc Laidlaw.

Read more of this story at Slashdot.

full article

Nov 19

Xiaomi today unveiled a new iteration of its virtual assistant Xiao Ai and shared a new feature of Android-based MIUI operating system as the publicly listed Chinese technology group pushes to expand its internet services ecosystem. From a report: At its annual Mi Developer conference in Beijing, the company said it is integrating an earthquake warning function into MIUI for select users in China, with plans to expand it nationwide soon. The integration, touted as the first of its kind globally, will enable alerts to be sent to smartphones running MIUI 11 and Mi TV “seconds to tens of seconds” before the quake waves arrive, Xiaomi said. The feature, which was first tested in September this year, has been developed in partnership with Institute of Care-life, a Chengdu-based organization focusing on natural disaster warning. Xiaomi said it has activated the feature for the earthquake-prone Sichuan Province and plans to expand it elsewhere in the nation soon. Wang Tun, head of the institute, said this function, unlike those available through apps in some countries, works more efficiently and does not rely on a working internet connection.

Read more of this story at Slashdot.

full article

Nov 19

samleecole writes: Using a photograph to algorithmically generate a person’s face and some 3D-rendered body parts, anyone can make a realistic avatar of someone who is walking around in real life. Import the avatar into another program, and you can have sex with them in virtual reality, without the real person ever giving consent. On forums like Reddit, marketplaces like Patreon, and on standalone websites, communities of anonymous users are making and selling computer-generated likenesses of celebrities and other real people. The 3D models that emerge from these communities can be articulated into any position, animated, modified, interacted with in real time, and manipulated in ways that defy the constraints of physical reality.

Like deepfake videos traded online, the technology to create photorealistic 3D models of real people is rapidly approaching — and it’s getting easier for the average user to access those tools and programs. Rendering a realistic human is a process which historically required the specialized technical knowledge of teams of artists in game and special effects studios. Those studios, traditionally, have to obtain the rights to use someone’s likeness before rendering them, but many hobbyists seemingly make avatars of anyone, with or without their consent.

Read more of this story at Slashdot.

full article

Nov 17

“Mozilla is no longer fighting for market share of its browser: it is fighting for the future of the web,” writes the Guardian, citing Mozilla Project co-founder Mitchell Baker:

Baker’s pitch is that only Mozilla is motivated, first and foremost, to make using the web a pleasurable experience. Google’s main priority is to funnel user data into the enormous advertising engine that accounts for most of its revenue. Apple’s motivation is to ensure that customers continue to buy a new iPhone every couple of years and don’t switch to Android….”

Firefox now runs sites such as Facebook in “containers”, effectively hiving the social network off into its own little sandboxed world, where it can’t see what’s happening on other sites. Baker says: “It reduces Facebook’s ability to follow you around the web and track you when you’re not on Facebook and just living your life….” Mozilla has launched Monitor, a data-breach reporting service; Lockwise, a password manager; and Send, a privacy-focused alternative to services such as WeSendit. It’s also beta-testing a VPN (virtual private network) service, which it hopes to market to privacy-conscious users…

Apple’s iOS (mobile operating system) is an acknowledged disaster for Mozilla. Safari is the default and, while users can install other browsers, they come doubly hindered: they can never be set as the default, meaning any link clicked in other applications will open in Safari; and they must use Safari’s “rendering engine”, a technical limitation that means that even the browsers that Firefox does have on the platform are technically just fancy wrappers for Apple’s own browser, rather than full versions of the service that Mozilla has built over the decades… “Even if you do download a replacement, iOS drops you back into the default. I don’t know why that’s acceptable. Every link you open on a phone is the choice of the phone maker, even if you, as a user, want something else.”

Summarizing Baker’s concerns, the Guardian writes that “It is perfectly possible to build a browser that prevents advertising companies from aggregating user data. But it is unlikely that any browser made by an advertising company would offer such a feature…”

And an activist for the Small Technology Foundation tells them that Google “wants the web to go through Google. It already mostly does: with eyes on 70% to 80% of the web.”

Read more of this story at Slashdot.

full article

Nov 17

Gizmodo reports that Sony “will happily sell you make-believe virtual meals” for their robotic Aibo dog to unlock tricks, one of several new features added since its re-launch in 2017:

The new feature that will appeal to most owners, however, is Aibo Food, which allows the robot to be virtually fed using augmented reality through the Aibo smartphone app. Meals can be purchased using coins, which are awarded to users through random actions like repeatedly using the Aibo app, or during special events. But once users runs out of coins, which is bound to quickly happen as they try out the new Aibo Food feature, they can either wait for more Sony handouts or purchase additional coins for a fee.

Sony points out that Aibo’s performance and features aren’t dependent on whether the dog is regularly fed — it is, after all, just a robot. So hopefully the company won’t change its mind down the line, making your pup act sluggish and distracted when you’re not forking out for pretend food…. Of course, other complications arrive once you start feeding an animal, and the new software update also allows users to finally potty train their Aibos using a new mapping feature so the robot doesn’t pretend-shit all over your house.

This appears to be a free feature, until Sony realises it can sell owners virtual poop bags.

There’s also a new web-based API/developer program that lets you program the robot dog to perform custom actions — and Aibo dogs now come equipped with some new patrol/security functionality.

“Using its facial recognition and room-mapping capabilities, Aibo will be able to patrol homes and locate various family members, providing reports on where everyone is, and helping owners track down specific people, according to Sony.”

Read more of this story at Slashdot.

full article

«     |     ?     |     »