Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning

iStock // Ekaterina Minaeva
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Apple Wants to Show Off Your Best Night Mode Photos as Part of a New Campaign

Austin Mann, Apple
Austin Mann, Apple

Calling all aspiring photographers who nabbed an iPhone 11 for the express purpose of trying out its fancy camera capabilities: It’s time for your night mode photos to see the light of day.

As Travel + Leisure reports, Apple is currently hosting a competition to find the best night mode photos taken on an iPhone 11, iPhone 11 Pro, or iPhone 11 Pro Max. You can submit your photos through January 29, after which a carefully selected team of experts will evaluate all submissions and announce the five winning images on March 4.

Judges include Arem Duplessis, the former design director of The New York Times Magazine; Darren Soh, an award-winning photographer from Singapore; Tyler Mitchell, the first black photographer to shoot the cover of American Vogue (his subject, rather memorably, was Beyoncé); and several other esteemed members of the industry.

golden gate bridge shot on iphone 11
The Golden Gate Bridge, shot on an iPhone 11 Pro.
Jude Allen, Apple

In addition to appearing on Apple’s homepage and Instagram (which has more than 21 million followers), the photos could also be featured in digital campaigns, Apple stores, third-party photo exhibitions, or even on physical billboards. In addition to all the exposure, the winners will be paid a licensing fee in exchange for granting the company complete freedom to use their work for one year.

To submit your shots, you can either share them on a public Instagram, Twitter, or Weibo account with the hashtags #ShotoniPhone and #NightmodeChallenge, or email your images to shotoniphone@apple.com—just be sure to title your files in this format: ‘firstname_lastname_nightmode_iPhonemodel.’

If you’re new to the iPhone 11 and aren’t quite sure how to snap photos in night mode, it’s easier than you might realize. The feature comes on automatically in dim or dark places and decides on a capture time for you (which you can always adjust). And if you think editing your photos afterward will increase your chances of winning the competition, that’s fine, too: Apple will accept photos edited in the app or even with non-Apple software.

You might want to avoid capturing the Eiffel Tower after dark, however—here’s why.

[h/t Travel + Leisure]

This Smart Speaker Is Designed to Help You Learn Chinese

Maybe
Maybe

This latest in smart speaker technology isn’t designed to turn on your Wi-Fi-connected lights or help you figure out a wine pairing. Lily, a new smart speaker now available on Indiegogo, can help you learn Chinese writing and speaking “faster than anything on the market,” according to its creators.

The voice-controlled gadget is equipped with artificial intelligence that allows it to engage in real-time conversations, so beginning to advanced students can learn to speak the Chinese language Mandarin. It can correct pronunciation, perform translations, and help you learn vocabulary through interactive games, too. It helps you learn to write with an associated app where you can practice using both Pinyin and Chinese characters. Lily’s curriculum can also prepare you for the HSK Chinese proficiency exams necessary to work or study in China.

A Lily speaker conversation illustrated with text bubbles that read 'Hey Lily, how do I introduce myself?' and '你好,你叫什麼名字'
Maybe

Designed by the language-learning company Maybe—which is based in San Francisco and Shenzhen—with Chinese tutors, the AI is engineered to recognize different accents, even if your pronunciation isn’t perfect.

Language learning is all about repetition, and unless you live in an area with a large Chinese population, you may not have much opportunity to practice your Mandarin in everyday life. A smart speaker like Lily can provide the next best thing by creating an immersion environment for you at home, putting it a step ahead of other language-learning apps like DuoLingo.

It’s currently only in the prototype stage, but it’s scheduled to ship by April 2020. As for other languages, the Lily team says that they hope to release software updates in the future that could give the speaker French and Spanish capabilities. (Once you have the speaker, you’ll be able to download any other languages that are released—you won’t have to buy a whole other language-specific speaker.)

The smart speakers start at $229 for early-bird buyers, which may seem like a significant investment, until you start looking at the prices of even a few weeks of private language classes. Get yourself one on Indiegogo. And don't worry about it matching your home decor—it comes in red, white, black, blue, teal, and pink.

Mental Floss has affiliate relationships with certain retailers and may receive a small percentage of any sale. But we only get commission on items you buy and don’t return, so we’re only happy if you’re happy. Thanks for helping us pay the bills!

SECTIONS

arrow
LIVE SMARTER