A nerdy father of two, a husband of a beautiful and understanding wife, an engineer who loves anime and tinkering with PCs and games, but most of all, loves God.
MindMaze’s Mask device includes a foam face pad embedded with electrodes that can capture facial muscle movements.
Tej Tadi is really excited about the foam face pad that sits inside a virtual-reality headset on a desk in his office.
It’s the kind of mundane piece of cushioning normally found on many kinds of head-mounted displays, but the one his startup, MindMaze, has is different. It’s embedded with electrodes so it can pick up the electrical signals associated with your facial muscle movements, and it’s connected to a computer where software analyzes these signals to reproduce your facial expressions on an on-screen avatar.
Called Mask, Tadi sees it as a way to bring natural-looking grimaces, smiles, and eyebrow raises to virtual characters without adding much bulk to headsets. Making it easier for users to express emotions—and interact with each other—in virtual reality could encourage more people to try it out, he thinks, and make it more effective.
“The only way to do it is to bring emotions back into the game,†he says. “That makes us human, right? The non-verbal cues.â€
For now, virtual and augmented reality headsets are still a tough sell to most consumers for many reasons: they’re annoying to wear, don’t yet have many clear practical applications, and can feel isolating, to name just a few.
Switzerland-based MindMaze, which has raised $100 million, already offers virtual-reality hardware that combines features like games, motion tracking, and brain-signal monitoring to help rehabilitate stroke victims; it’s currently used in several dozen hospitals in Europe.
Mask builds on the company’s existing work, Tadi says, and can currently recognize 10 different expressions, including winking, smiling, smirking, grimacing, and eyebrow-raising. And with a microphone attached to it, it can mimic the wearer’s mouth while they’re speaking, too.
Tadi expects such a gadget will be available to consumers later this year, either as a product from MindMaze itself or from a headset maker.
As I watched a member of the MindMaze team try out a prototype of the device, it seemed to work quite well: with an OSVR virtual-reality headset outfitted with MindMaze’s technology, he made a variety of expressions that a cartoony male character also made on a desktop display. The speaking mimicry didn’t seem much better than basic “talking†motions I’ve seen on virtual characters’ mouths, however.
There was no need to calibrate the headset for me to try it, but it had issues recognizing some of my facial expressions. Tadi and his team explained that this may have been due to stray hairs getting in the way of one of the electrodes. Â
And it’s still unwieldy and messy: the electrode-laden foam insert was connected to some wires and electronics as well as to a computer, and a gel-swabbed electrode was connected to each of my earlobes to serve as a reference.
Tadi says the ear electrodes could be replaced by dry ones that are embedded in headphones connected to the headset, and that the electronics could be whittled down. But the fact remains that the feature would be adding even more baggage to headsets that are already criticized for their size—it remains to be seen whether the addition of emotional expression can outweigh that issue.
from Technology Review Feed – Tech Review Top Stories http://ift.tt/2otbsrv
via IFTTT
After years of ignoring streaming services, Cannes is finally recognizing both Netflix and Amazon’s original work. With several Amazon funded-movies debuting at last year’s event, Cannes’ 2017 programme reveals that Netflix original movies will also finally be joining the festival. Netflix will be premiering The Meyerowitz Stories at the festival, starring Adam Sandler and Ben Stiller. It will also be debuting Okja for the first time, Netflix’s new Korean-directed monster movie featuring Tilda Swinton and Jake Gyllenhaal. Amazon movies will be featured for a second year, with Wonderstuck andYou Were Never Really Here getting first screenings.
Screening movies funded by streaming sites isn’t the only notable change for Cannes’ 17. This year’s festival also steps away from cinema for the first time in its history, choosing to also recognize television. Alongside the big names in cinema, Cannes will also debut the first two episodes of Twin Peaks‘ long-awaited new season. All six hours of Jane Campion’s Top of the Lake season 2 will also be screened at the event for the first time.
This will be the first time the international film festival has recognized streaming services, showing a marked shift in how the world consumes cinema. Whether this is simply because of the caliber of directors Amazon and Netflix attract rather than the services remains to be seen, but with more and more Hollywood talent flocking to streaming services – they’re becoming increasingly difficult to ignore.
According to Telsa CEO Elon Musk, the automotive arm of his company will unveil a semi truck this coming September. Musk dropped the news via a tweet. The company is currently gearing up to start full production of the Model 3, set to start being delivered at the end of 2017. But now it looks like Musk is getting ready to dive into into long-haul trucking.
Tesla Semi truck unveil set for September. Team has done an amazing job. Seriously next level.
Musk talked about the semi truck and a minibus last year during an earnings call. It’s actually not too surprising that Tesla would get into the semi-truck market. Mercedes and Otto (which was acquired by Uber) both demonstrated the need for autonomous and electric trucks. Tesla making its own model is the next logical step for the company.
Sixty million dollars. That’s roughly how much it costs to send a payload to orbit on SpaceX’s Falcon 9 rocket. It’s actually a bargain for space, but it’s far more than University of Central Florida physicist Julie Brisset, who seeks to study the early stages of planet formation in microgravity, can afford. She’s one of countless researchers unable to access the low gravity laboratory of outer space because of the staggering price tag.
That’s why Brisset is eyeing New Zealand-based startup Rocket Lab with excitement. The company, which was founded in 2006, hasn’t flown a balloon into orbit yet. Its Electron rocket has a maximum payload capacity of just under 500 pounds, which is peanuts compared with the Falcon 9’s 50,265-pound weight limit. But Rocket Lab does have one big thing going for it: It wants to launch rockets into space for just five million dollars.
If the company can pull that price point off—still a big ‘if’ as it hasn’t begun commercial flights yet—five million dollar launches would be a game-changer, especially for folks operating small, lightweight satellites like CubeSats. And the small satellite market is Rocket Lab’s exclusive focus. “Launch has become a bottleneck for many satellite companies, and Rocket Lab aims to solve that by providing unprecedented access to space,†Peter Beck, cofounder of Rocket Lab, told Gizmodo.
CubeSats are a type of “nanosatellite,†with a maximum weight of roughly 2.93 pounds, that have a standard size—10 x 10 x 11.35 cm, or some larger multiple of those dimensions. The CubeSat design dates back to 1999, when researchers at California Polytechnic State University and Stanford sought to provide graduate engineering students a cheap and feasible way to experiment with spacecraft and send them into orbit. Over time, the design has become an industry standard.
“The really revolutionary aspect of CubeSats is their small format,†Brisset, who is working on multiple CubeSat experiments, told Gizmodo. “This makes CubeSats very cost-effective in fabrication and transportation to space, and it significantly reduces their manufacturing time.†A full-sized spacecraft can cost an enormous amount of money, in the range of several hundred million dollars, Brisset explained. But if you don’t need the complexity and durability that a large spacecraft affords—if you just need to send a single scientific instrument into space for a year, for instance— you might turn to nanosatellites instead.
In 2011, there were 12 total nanosatellite launches. In 2017, there are a projected 569. From package tracking to satellite Internet to studying the impacts of climate change, the ability to monitor the Earth from space is a benefit to small and large companies and governments—and CubeSats are making that possible. At least, for those who can shore up the $100,000 needed to stick one of these three-pound hunks of metal in the cargo capsule of a Falcon 9 rocket or similar launch vehicle.
Cost is a big impediment for the small satellite industry, but it’s not the only one. Small satellites operators are also constrained by payload capacity, forced to hitch a ride on existing launches that have extra room. As result, there’s a severe backlog, and the projects that make the cut tend to be the ones with ties to government agencies like NASA. For example, AOSAT-I, or the Asteroid Origins Satellite, is a three-unit CubeSat launching later this year that’s designed to simulate the surface of an asteroid to help NASA refine the design of future asteroid landers. Then there’s Mars Cube One, or MarCo, a fleet of CubeSats that will be deployed in Martian orbit to function as a communications relay for NASA’s InSight Mars lander. These projects are undoubtedly exciting—but for every small satellite with a ticket to orbit, there are dozens more on the waiting list.
Instead of treating small satellites as an afterthought, Rocket Lab is centering on these projects, and the companies who want to launch them affordably. Its Electron rocket, whose Rutherford engine is constructed with 3D-printed carbon composite parts, is designed to keep down costs. “We chose 3D printing as it lends itself to rapid manufacturing, in turn reducing the cost and time of creation and increasing our ability to launch with higher frequency,†said Beck. The rocket is also unique in that it claims the first-ever partially battery-powered engine—it uses electric motors that run on lithium batteries to pump liquid fuel into a combustion chamber, eliminating the need for heavy turbopumps, and additional fuel to operate them. While the design is not yet flight proven, Beck told The Verge electric batteries reduce the cost and overall complexity of the engine.
If you take Rocket Lab’s $5 million, 500-lb capacity, the price per CubeSat comes out to roughly $30,000. If a company wanted to buy out the entire launch vehicle, that might also be feasible. “There’s a lot of advantages that can come from a smaller launch provider, where we can afford to buy the full rocket, and then we have a lot more say in terms of exactly what orbit we’re going into, when it’s going, what safety requirements we have to adhere to,†Mike Safyan, Director of Launch and Regulatory Affairs at private satellite operator Planet, told Gizmodo.
Rocket Lab not the first company to seek to provide for the small satellite market—Vector Space Systems and Virgin Orbit are two others that come to mind—but it’s one of the furthest along. Recently, the company announced that its Electron rocket is ready for flight tests, the first of which should occur within the next few months.
It’s worth taking any company’s claims of a revolutionary new technology with a grain of salt. But at least a few others in the space industry, including Phil Larson, former SpaceX communications manager, and currently with the University of Colorado, Boulder, are optimistic about Rocket Lab’s chances of achieving its $5 million price point. “Rocket Lab is [one of the companies] innovating and helping encourage a whole new smallsat industry,†Larson told Gizmodo. “What they’re attempting to do is definitely ambitious, but I think it’s achievable as well—and would do wonders to help more of these smallsat startups close their business case.â€
Brisset also has been excited to follow Rocket Lab’s journey. Her team envisions CubeSats as a platform for countless microgravity experiments, including her own project, which studies the early stages of planet formation.
In February, India launched 104 satellites on one rocket, shattering all previous records. Eighty-eight of those small satellites were the property of Planet; with these eyes on the sky, along with the 50 they already had in orbit, the company promises its customers high-resolution images of the Earth for everything from crop yield monitoring to aiding first responders with real-time images of natural disasters. “We have 132 [small satellites] that are actively imaging right now from space,†Safyan said. “When the [Oroville Dam] broke in Northern California, we had some really interesting imagery to see the impact of that and flooding in local areas.â€
Satellites are already changing the way we see our planet. If companies like Rocket Lab are successful in driving down costs enough to offer a new generation of researchers a ticket to orbit, they might just change the way we see the entire universe.
Swapna Krishna is a space science and tech writer and writes a weekly space column called Space Matter at Paste Magazine.
Boeing expects to shave $2 to $3 million off each 787 Dreamliner’s manufacturing costs by 2018, thanks to 3D-printed titanium. The company has teamed up with Norwegian company Norsk Titanium to create the first printed structural titanium components for a plane. As Reuters notes, General Electric already prints fuel nozzles for aircraft engines. However, this is the first time a company is using 3D-printed components for parts of a plane that bear the stress of an airframe during a flight.
Boeing turned to 3D printing for the 787, because it requires more metal than its other models. Plus, traditionally manufactured titanium alloy can be very expensive, especially since the company makes 144 Dreamliners a year. The aerospace corporation’s partnership is a resounding recommendation for printed metals in the aviation industry and is proof that companies are starting to trust the manufacturing process and its resulting materials.
From early 2016 to February 2017, Boeing worked with Norsk to be able to pass the Federal Aviation Administration’s rigorous testing program for the components. The partners expect to get additional FAA approval for the material’s properties and manufacturing process later this year. That will allow the Norwegian firm to make more 3D-printed titanium parts without having to get each of them approved, leading to even more savings per plane.
Google wants to help you get in touch with your inner Picasso. Today, it’s launching AutoDraw, a web-based tool that uses machine learning to turn your hamfisted doodling into art. It’s similar to, but clearly far more advanced than, Android Wear’s ability to recognize a crudely drawn smiley face and replace it with an emoji.
The app is free and it works on any phone, computer or tablet. It’s pretty straightforward: draw your best version of a cake, for example, and the auto suggestion tool will try to guess what that amorphous blob actually is. Then, you can choose from a number of better looking cakes made by talented artists. Or, if amorphous blob is actually what you were striving for, you can turn off the auto suggestions and doodle away.
AutoDraw uses the same technology as another Google experiment called QuickDraw. It’s a mini-game that tells you which objects to draw, like an eye or a helicopter, then gives the AI 20 seconds to recognize it. AutoDraw is more of a creative tool, allowing users to make things like posters or coloring books. But, both likely serve the same purpose of teaching a neural network to recognize doodles.
Right now, Google claims AutoDraw can guess hundreds of drawings, and the company plans to add more in the future. If you have suggestions on what objects Google should add, or you’re an artist who’d like to contribute to the project, you can do that here.
Google has announced that its widely used Octane JavaScript benchmark is being retired, with Google saying that it’s no longer a useful way for browser developers to determine how best to optimize their JavaScript engines.
Octane was developed for and by the developers of V8, the JavaScript engine used in Chrome. It was intended to address flaws in the earlier SunSpider benchmark, developed by Apple’s Safari team. SunSpider’s tests were all microbenchmarks, sometimes testing something as small as a single operation performed thousands of times. It wasn’t very representative of real-world code, and it was arguably being gamed, with browser vendors introducing optimizations that were aimed primarily, albeit not exclusively, at boosting SunSpider scores. This was being done even when those optimizations were detrimental to real-world performance, because having a good score carried so much prestige.
Octane was introduced in 2012 and includes cut-down versions of somewhat realistic workloads, such as compiling the TypeScript compiler. But since then, JavaScript coding styles have changed. JavaScript itself has changed; ECMAScript 2015 (the standardized version of JavaScript) introduced a range of new features that weren’t available in 2012, and hence aren’t tested by Octane, and all manner of new libraries and frameworks have emerged since.
At first Octane provided useful focus for the engine developers, highlighting areas that needed improvement. But, just as with SunSpider before it, Google has found that optimizations have been developed to boost Octane even if it hurts other scenarios. Once again, the desire to get the highest score possible has come at the expense of developing a better scripting engine. With all browsers now fast at Octane—Edge is a little ahead of Chrome, which is a little ahead of Firefox—Google has chosen to retire the benchmark.
This habit of gaming benchmarks into uselessness is as old as benchmarking itself. Some benchmarks, such as the SPEC CPU integer and floating point benchmarks, have rules for which compiler optimizations are permitted; they have to be applicable to “a class of problems […] larger than the SPEC benchmarks themselves” in an attempt to prohibit compiler vendors including optmizations that are good for SPEC and nothing else. This has not prevented extremely specific optimizations being used in the past. But browser benchmarks, which have no rules on which scores are and aren’t “official” lack even this kind of control.
The best protection against this kind of gaming is to develop better benchmarks. Google has developed infrastructure to allow it to repeatably and consistently time the loading of entire Web pages (thus covering not just JavaScript performance but also HTML and CSS performance) which it uses to measure real-world performance on 25 popular sites and guide development. The Safari developers created JetStream to replace SunSpider, containing a wider mix of tasks and a greater proportion of real applications. The Speedometer benchmark, also from WebKit/Safari developers, is artificial, but Google has discovered that it corresponds well to real-world performance of popular sites.
In spite of the problems, the desire to benchmark and have repeatable, objective measures of performance won’t go away. As new benchmarks are developed, we’d expect the cycle to repeat itself; first they provide a useful target for improved performance, but then they become the primary goal. Real-world testing of the kind Google performs acts as a useful backstop, discouraging the company from doing anything that’s detrimental to the Web at large, but the incentive to skew things will never disappear completely.
