Monthly Archives: March 2018

The Bionic(le) Man: Kids literally arms himself via Lego

Weak side: Complaining about doing push-ups.
Strong side: Being born with a partial arm, saying F it, Imma build myself a Lego arm & do push-ups on that

Ever since he was a kid, David Aguilar was obsessed with Lego. He spent his childhood building cars, planes, helicopters, and eventually, his own prosthetic. Born with a deformed arm, the self-named “Hand Solo” decided to take his Lego-building skills to the next level. At age 18, he perfected his designs with the MK2, a prosthetic arm with the ability to bend and pick up objects with a pincer-like grip. Now, he’s the coolest kid on the block.


[Vimeo] [Via Maria Brenny]

The terrific behind-the-scenes making of Apple’s “Welcome Home”

Spike Jonze, a quarter century on from Sabotage (Dios mio…), continues to crush it. AdWeek writes,

The almost seven-minute mini-documentary, directed by Danilo Parra, explores all the creative disciplines in wonderful detail, shedding light on the challenges that the intricate piece presented every step of the way.

And if you somehow haven’t seen the finished piece, enjoy:

[YouTube 1 & 2]

A beautiful visualization of flight paths in AR

It’s a running joke at Google that you can spend your whole career not knowing what the people nearby you are doing, or even who they are. (In fact, I continue to harbor a dream about creating an AR overlay that would solve this, but that’s another story.) As it happens, I just discovered that my teammate Bill has been making some really cool augmented reality experiments, including the open source Flight Paths:

Flight Paths is an experiment that transforms your room into a flight path visualization. Touch any horizontal surface and explore as flights take off from JFK or SFO and fly around your space. Learn more at

Continual note to self: Ask people more about what they’re doing!



New eerily realistic “Siren” performance capture & rendering

We now find ourselves in one of those internal arms races, with one group of nerds working to realistically smooth real people’s skin in realtime while others work to make fake people’s skin realistically imperfect. ¯\_(ツ)_/¯

‘Siren’, a high-fidelity, real-time virtual “human,” is the brainchild of Epic Games, 3Lateral, Cubic Motion, Tencent, and Vicon, and was designed as a more effective, timesaving alternative for creating hyperrealistic video game characters… Look up close, and you’ll find that ‘Siren’ even has beauty “imperfections” like pimples and visible pores.


[YouTube] [Via]

1,760 LED screens dance in Coke’s epic 3D billboard

A new drink awaits you in the off-world colonies! A chance to imbibe again…

Atlanta Inno writes,

Coca-Cola unveiled the six-story digital advertisement in August, winning two Guinness World Records titles: “The Largest 3D Robotic Billboard” and “The First 3D Robotic Billboard.” The company has featured an advertisement in Times Square for more than 97 years and the latest sign features 1,760 independently moving LED screens, creating a multisensory experience for the audience, according to a news release.



Try Google Lens inside Google Photos

Now available on both iOS & Android, and offering a few neat tricks:

Lens works on photos of business cards, books, landmarks and buildings, paintings in a museum, plants or animals, and flyers and event billboards. When you use Lens on a photo that has phone numbers or an address, you can automatically save this information as a contact on your phone, while events will be added to your calendar.





Google Maps + Unity FTW!

Get ready for a whole new wave of AR gaming:

Per The Verge,

Unity integration will also allow developers to customize maps with what appears to be a great deal of flexibility and control. Things like buildings and roads are turned into objects, which developers can then tweak in the game engine. During a demonstration, Google showed off real-world maps that were transformed into sci-fi landscapes and fantasy realms, complete with dragons and treasure chests.

Jacoby says that one of the goals of the project was to help developers build detailed worlds using Maps data as a base to paint over. Developers can do things like choose particular kinds of buildings or locations — say, all stores or restaurants — and transform each one. A fantasy realm could turn all hotels into restorative inns, for instance, or anything else.



Google makes an AI-powered musical instrument (for real!)

Expressive superpowers FTW!

The Verge writes,

The NSynth Super is a piece of hardware that brings the NSynth technology to life. As Douglas Eck, research scientist on the Google Brain team, says in the video above, NSynth doesn’t generate notes, but rather, the actual sound of an instrument. The NSynth algorithm learns the core qualities of what makes up an individual sound and then is able to combine sounds to create something completely new.


[YouTube 1 & 2]

The secrets behind rock-solid microvideos on Pixel 2

Continuing our series of Research Blog posts (see realtime segmentation, motion tracking), my teammates have provided an inside look at the tech they’ve developed—this time covering how motion photos get stabilized on the fly:

By combining software-based visual tracking with the motion metadata from the hardware sensors, we built a new hybrid motion estimation for motion photos on the Pixel 2. 

Check out the blog post for details, or just enjoy lots of good before/after examples of stabilization in action.


Check out Kittyhawk’s autonomous electric flying taxi

Yeah, if I had Google-founder money, I’d be funding craziness like this pretty much nonstop. Fortunately, actual Google founder Larry Page is doing just that.

TechCrunch writes,

With Cora, Kitty Hawk’s real vision comes through – the startup aims to build, own and operate a fleet of these as an on-demand self-flying taxi service, beginning commercial operations in as few as three years’ time. No word yet on price or how this will operate exactly in terms of pick-up and drop-off points, but it seems farther along than a lot of the other pie-in-the-sky autonomous aerial taxi projects out there, thanks to official government support.



Google “Pup View” comes to Japan

Yo dawg, I put a GoPro on your dog…

Speaking of Google’s why-the-heck-not streak:

Akita dogs are a big part of life in Ōdate city. So to give visitors a snapshot of the local area, residents called on the help of a hardy trio of Akita dogs—Ako, Asuka and Puuko—to give their unique perspective of a dog’s life in Akita. With a small camera attached to a dog-friendly harness on their backs, the dogs set out to collect their very own Street View Pup-View of their home city.



“Let’s-A go!” Mario Kart comes to Google Maps

Happy “Mar10” Day! All this week you can have a little Italian plumber & his kart drive along with you:

To get started, you’ll need to first update your app from Google Play or the App Store. Next, simply click on the yellow “?” icon found on the bottom right of your Google Maps app on Android or iOS. You’ll then see a prompt to enable Mario Time!

Whenever I see these little “why the heck not?” bits from Google, I think of this observation from Khoi Vinh, occasioned by the Smarty Pins map/trivia game:

Apple fans like myself often criticize Google for doing things that Apple would never do, and Smarty Pins is a prime example of that. Aside from being an unfair criticism, it’s pointless. The fact that Google endeavors to produce silly things like this is on the whole a positive thing, I believe. It’s acting according to its own compass, which is what every company should be doing.

At Adobe I used to say, “We’ll never out-Apple Apple. We’ll never be more mysterious & magical, so let’s be ourselves—conversational and down-to-earth.”

Or as Steve Jobs said, “Your time is limited, so don’t waste it living someone else’s life.”


Google AI powers… Cheetos Vision?

As I’m wont to say ~15 times daily, ¯\_(ツ)_/¯.

Developed by the GS&P Labs at Goodby Silverstein & Partners, the AI builds upon Google’s open-source ‘Tensorflow’ machine learning framework to create a photography app that packs a crunch. The AI is trained to closely replicate the irregular silhouettes of Cheetos bits to fill in images completely—none of that bag-of-air nonsense.


[YouTube] [Via]

Engineers & testers: Come work on Photoshop

Turns out that my Adobe friends are hiring for a lot of positions in & around Photoshop engineering. If any of the following sound like an interesting match for your skills, let ‘em know!









[Via Jackie Lincoln-Owyang]

Google introduces Flutter, a cross-platform UX library

Check it out:

As Ars Technica explains,

Flutter apps don’t directly compile to native Android and iOS apps; they run on the Flutter rendering engine (written in C++) and Flutter Framework (written in Dart, just like Flutter apps), both of which get bundled up with every app, and then the SDK spits out a package that’s ready to go on each platform. You get your app, a new engine to run the Flutter code on, and enough native code to get the Flutter platform running on Android and iOS.

Also, I’m totally creating a band called Stateful Hot Reload. 🙂



Microsoft’s 3D Soundscape app helps blind users navigate the world

Augment all the humans! Check out this new perceptual enhancement:

The app, Soundscape, calls out roads and landmarks as they’re passed, and lets users set audio beacons at familiar destinations. If at any time you’re unsure of where you are, or which direction to head in, you can simply hold the phone flat in your hand and use the buttons on the bottom of the screen to locate nearby roads and familiar destinations.



Kids in Africa experience Legos for the first time

Seeing this & remembering how wildly creative I’ve seen little kids be when given just a handful of wheels & bricks brought a big smile to my face:

Support our Crowdrise campaign! We sent 50 pounds of Legos to a school in a rural Uganda. This video captures the students, none of whom had ever seen Legos before, playing the Legos for the firs time. For more information, visit


[YouTube] [Via Sally Cox]

Check out our realtime video segmentation

When I learned After Effects in ye Olden Times (aka the Carter—okay, Clinton—Administration), rotoscoping (selecting a portion of a scene over many frames) was such a brutally slow and painful process, it’d make a Photoshopper weep & wish to run home to the Pen tool. Over the years amazing tools like Roto Brush removed some of the drudgery, but the process has remained largely manual.

Now, though, machine learning can teach computers to perform this kind of segmentation automatically, in realtime, on a friggin’ telephone. Thanks largely to the efforts of the Belarusian team Google acquired in August, this happened:

Today, we are excited to bring precise, real-time, on-device mobile video segmentation to the YouTube app by integrating this technology into stories. Currently in limited beta, stories is YouTube’s new lightweight video format, designed specifically for YouTube creators. Our new segmentation technology allows creators to replace and modify the background, effortlessly increasing videos’ production value without specialized equipment.

So how does this witchcraft actually work? I’m so glad you asked: check out the details in this post on the Google Research blog. And stay tuned, as Teh ML Hotness is just getting warmed up.



Photography: Beautiful doom envelops flowers

Some say the world will end in fire,
Some say in ice…

Photographer Thomas Blanchard “represents all four seasons by showing flowers blooming, submerged in water that freezes over, burning, and shrouded in clouds of colorful inks.” Amazing (stick with it):

From what I’ve tasted of desire
I hold with those who favor fire.
But if it had to perish twice,
I think I know enough of hate
To say that for destruction ice
Is also great
And would suffice.

NewImage NewImage