On a drizzly recent Sunday night, the Art Deco lobby of 30 Rockefeller Plaza was clogged with celebrities attending "Saturday Night Live" 's fiftieth-anniversary show. Taran Killam, who played Donald Trump on the show in 2015, strolled in wearing a double-breasted tux and stopped cold when he noticed a curly-haired man in a green suit standing beside a rig that looked like a ten-foot-tall Swiss Army knife. "Oh, here it is," Killam said. The man was Cole Walliser, a forty-three-year-old director, and the gizmo was the Glambot, a high-speed camera mounted on a giant robotic arm. Walliser oversees the device during the E! network's red-carpet coverage, and the resulting dramatic slo-mo celebrity action clips are posted to TikTok and Instagram.

As I ran around the house picking up discarded cat toys and odd socks, I began to feel a nagging sense of doubt. I knew that no matter how hard I worked, I would soon receive a distress call from our robot vacuum, Gizmo IV, who always gets stuck in the same spot under the couch. After Gizmo finishes, I then have to pick hair out of the carpet, which Gizmo expertly clumps but never sucks in. After that, I could look forward to a face full of dust as I emptied its tiny, barely-half-full bin (despite Gizmo's protests that it can hold no more). In fact, the week after pitching this very article, I came downstairs to find foul-smelling skid marks across the floors. At first, I feared an infection had prompted her to wipe all over the house.

To do dense depth estimation like in that article, I have difficuty to show the 3D model generated into jupyter. So I decide to write the obj2html library, you can find the original code in this repo. This view is interactive and you can use mouse right click to pan and mouse left click to gizmo, you can also zoom with mouse wheel. With a 3D model saved as .obj The demo colab notebook is here.

One of Silicon Valley's million robot companies brought another cool toy to the IFA conference in Berlin this month, and it sees, hears, and speaks as it cleans. The company is Trifo: They've been around for a few years, and they're tapping into a seriously lucrative market. According to the company, Statista estimates that shipments of home robots will grow to nearly 30 billion units by 2025 and The International Federation of Robotics says the personal household service robots market will reach over $10 billion by 2020. The new gizmo, announced at the IFA event in Berlin last week, is Max, a delightfully simple brand for a pretty curious, aware, and able robot. Trifo also launched Ironpie at CES in Las Vegas early this year, a smart robot vacuum that reportedly "cleans faster, protects furniture better, is controllable from anywhere, and with a host of features that improve on the original home vacuum concept."

Google, which prides itself on developing simple, intuitive software that seems to know what you want almost before you do, is finding itself in a very different world when it comes to its own phones and other gadgets. Its new Pixel 2 phones, released in October, got high marks for their camera and design -- at least until some users complained about "burned in" afterimages on their screens, a bluish tint, periodic clicking sounds and occasionally unresponsive touch commands . Then the company's new Home Mini smart speaker was caught always listening . Finally, its wireless "Pixel Buds" headset received savagereviews for a cheap look and feel, mediocre sound quality, and being difficult to set up and confusing to use. In short, Google is re-learning an old adage in the technology business: Hardware is hard.

It seems like only a few years ago that we began making wry jokes about the doofus minority of people who walked down the street while texting or otherwise manipulating their phone, bumping into lamp-posts and so forth. Now that has become the predominant mode of locomotion in the city, to the frustration of those of us who like to get anywhere fast and in a straight line. Pedestrian accidents are on the rise, and some urban authorities are even thinking of installing smart kerbside sensors that alert the phone-obsessed who are about to step into oncoming traffic. New technologies, as Adam Greenfield's tremendously intelligent and stylish book repeatedly emphasises, can change social habits in unforeseen and often counterproductive ways. The technological fixes to such technology-induced problems rarely succeed as predicted either.

As I cruise down Silicon Valley's Bayshore Freeway, I twirl my right index finger in small clockwise circles. Though the gesture recognition system needs several tries to pick up the command and crank up the 16-speaker stereo, the 20 air chambers massaging my backside keep my frustration at bay. All the while, the 2017 BMW 5 Series is doing the driving, maintaining a safe speed and staying between the lane lines. And here, surrounded by an armada of tiny sensors and whiz-bang gizmos, I wonder: Has the Ultimate Driving Machine yielded to the Ultimate Autonomy, Luxury, and Technology Machine? In a former life, the 5 Series was the sports sedan, the car Audi and Mercedes-Benz engineers saw in their sleep.

JUST as native English-speakers stumble with Japanese, the Japanese struggle mightily with English, not to mention Korean or Chinese. In the widely used TOEFL tests of English as a foreign language, Japan invariably ranks second from bottom among the 29 countries participating in the scheme. Compared with the 150m people around the world who speak English as a second language, there are only 9m non-native speakers of Japanese--and most of those were forced to learn the language during Japan's era of colonial occupation, and are now dying of old age. For those who put their faith in technology, therefore, it was encouraging to hear Shinzo Abe, Japan's prime minister, demonstrate his linguistic skills a few weeks ago with a palm-sized gizmo that provided instantaneous translations of spoken Japanese into near-flawless English and Chinese. Mr Abe can manage perfectly well without such a device, being one of the few Japanese prime ministers in recent years to speak English fluently.

They might not be delivering our mail (or our burritos) yet, but drones are now simple, small, and affordable enough that they can be considered a toy. The Parrot AR Drone has an API that lets you control not only the drone's movement but also stream video and images from its camera. In this post, I'll show you how you can use Python and node.js to build a drone that moves all by itself. So given that I'm not a drone, or a machine vision professional, I'm going to have to keep things simple. For this project, I'm going to teach my drone how to follow a red object.