Hype Cycle : Machine Learning

😂

3D Printing A Fabulous Lion [x]

This teenager invented an AI-based app that can quickly diagnose eye disease

This ingenious system is called Eyeagnosis, and Kopparapu recently presented it at the O'Reilly Artificial Intelligence conference in New York.

Her name is Kavya Kopparapu and she’s a 16-year-old high school junior. She just might be a South Asian-American Bill Gates in the making. 

Launching the Future of Space Communications

Our newest communications satellite, named the Tracking and Data Relay Satellite-M or TDRS-M, launches Aug. 18 aboard an Atlas V rocket from our Kennedy Space Center in Florida. It will be the 13th TDRS satellite and will replenish the fleet of satellites supporting the Space Network, which provides nearly continuous global communications services to more than 40 of our missions.

Communicating from space wasn’t always so easy. During our third attempt to land on the moon in 1970, the Apollo 13 crew had to abort their mission when the spacecraft’s oxygen tank suddenly exploded and destroyed much of the essential equipment onboard. Made famous in the movie ‘Apollo 13’ by Ron Howard and starring Tom Hanks, our NASA engineers on the ground talked to the crew and fixed the issue. Back in 1970 our ground crew could only communicate with their ground teams for 15 percent of their orbit – adding yet another challenge to the crew. Thankfully, our Apollo 13 astronauts survived and safely returned to Earth. 

Now, our astronauts don’t have to worry about being disconnected from their teams! With the creation of the TDRS program in 1973, space communications coverage increased rapidly from 15 percent coverage to 85 percent coverage. And as we’ve continued to add TDRS spacecraft, coverage zoomed to over 98 percent!

TDRS is a fleet of satellites that beam data from low-Earth-orbiting space missions to scientists on the ground. These data range from cool galaxy images from the Hubble Space Telescope to high-def videos from astronauts on the International Space Station! TDRS is operated by our Space Network, and it is thanks to these hardworking engineers and scientists that we can continuously advance our knowledge about the universe!  

What’s up next in space comm? Only the coolest stuff ever! LASER BEAMS. Our scientists are creating ways to communicate space data from missions through lasers, which have the ability to transfer more data per minute than typical radio-frequency systems. Both radio-frequency and laser comm systems send data at the speed of light, but with laser comm’s ability to send more data at a time through infrared waves, we can receive more information and further our knowledge of space.

How are we initiating laser comm? Our Laser Communications Relay Demonstration is launching in 2019! We’re only two short years away from beaming space data through lasers! This laser communications demo is the next step to strengthen this technology, which uses less power and takes up less space on a spacecraft, leaving more power and room for science instruments.

Watch the TDRS launch live online at 8:03 a.m. EDT on Aug. 18: https://www.nasa.gov/nasalive

Join the conversation on Twitter: @NASA_TDRS and @NASALasercomm!

Make sure to follow us on Tumblr for your regular dose of space: http://nasa.tumblr.com

Only in Japan : Azuma, the First Hologram Communication Robot

disneyandmore reports:  Azuma is a cheerful young girl about twenty centimeters tall, with long blue hair, living under a glass bell. Lying near the sofa in the living room or on the night table, she welcomes her owner’s bachelor with a big smile when he comes home from work, then turns on the lights and informs him of the weather tomorrow. When her darling is not there, she sends him text messages to remind her that she thinks about him and is languishing. This kind of Tinkerbell is called Azuma Hikari, and it is a character in the manga style: diaphanous skin, interminable legs, huge eyes and childish voice. She is the first occupant of the Gatebox, the magic box launched in December 2016 by Japanese company Vinclu. Once activated, the translucent bubble, placed on a black base, lights up and Azuma appears inside, like a hologram, it’s the first “Hologram Communication Robot”, as they call her.

Azuma lives her life inside the transparent globe. She sleeps, brushes her teeth, takes a cup of tea sitting in an armchair, but rises as soon as her “master”, as named on the site of the apparatus, solicits her. Equipped with a camera and a microphone, she recognizes him, distinguishes his movements, understands a few words and interacts with him, like virtual assistants like Siri.

In a video of demonstration, she is seen under her dome, tiny and smiling, sitting next to her young master watching television. It looks like a couple would be lazing on a Sunday night. “It is a comforting character for those who live alone,” says Vinclu’s website, which defines this virtual girlfriend as gentle, considerate and playful. One understands the goal: this little fairy of the house is supposed to fight the loneliness in a country where the number of singles grows. With Azuma, promises the company, “you will finally want to go home early.”

Just created, its Artificial Intelligence (AI) currently only includes Japanese, and the Gatebox is only available in Japan and the United States. Only 300 copies have been sold to date - at the price of 298,000 yen (2,400 euros). Will this virtual girl become the miniature companion of men alone? But how does it works? Is it really a hologram? Well, not really, it is actually a projection on a glass plate and it uses the old and very famous “Pepper Ghost effect”, the same one that WDI Imagineers use in Disney’s Haunted Mansion since 50 years!

“Of course machines can’t think as people do. A machine is different from a person. Hence, they think differently. The interesting question is, just because something thinks differently from you, does that mean it’s not thinking ?”

- The Imitation Game

Uncanny Rd

Project by Anastasis Germanidis and Cristóbal Valenzuela is an online Neural Network semantic painting tool. Trained on visual road data, you can doodle or place preset markers which will be visually translated into another image:

Uncanny Road is an experimental tool for collectively synthesizing a never-ending road using Generative Adversarial Neural Networks. It is based on the pix2pixHD project, published by @nvidia and UC Berkeley (Project, Paper and Code), that allows for photorealistic image-to-image translation. The pix2pix model was trained using adversarial learning on the Cityscapes dataset, containing thousands of street images.

To synthesize street images, draw on the colormap of the scene. Each color represents a different kind of object label (e.g. road, building, vegetation, etc.) that the neural network can understand

Try it out for yourself here

Bitcoin micropayments: Coming to smartphones—and toasters?—everywhere

Micropayments might not top your list of most compelling inventions, but they’re a sought-after capability. Small payments of less than a dollar, or even less than a cent, have the potential to shake up old, established business models, and open up new doors for the Internet of Everything.

Small digital payments have been tried again and again—in fact, Web inventor Tim Berners-Lee tried to embed micropayment capability into the original World Wide Web, but without success. So far, inherent transaction costs have been an unsurpassable hurdle.

Some argue that digital payment methods like bitcoin are the way forward.

继续阅读

Example-based Face Stylization - Web Demo

DCGI and Adobe Research have put up an online interactive demo of their stylized facial animation paper.

Just drag and drop an image with a face into it, select one of the styles on the right, hit ‘Submit’ and see what happens …

Try it out for yourself here