The Tens Of Millions Of Faces Training Facial Recognition; You’ll Soon Be Able To Search For Yourself

August 25, 2019 by 46 Comments

In a stiflingly hot lecture tent at CCCamp on Friday, Adam Harvey took to the stage to discuss the huge data sets being used by groups around the world to train facial recognition software. These faces come from a variety of sources and soon Adam and his research collaborator Jules LaPlace will release a tool that makes these dataset searchable allowing you to figure out if your face is among the horde.

Facial recognition is the new hotness, recently bubbling up to the consciousness of the general public. In fact, when boarding a flight from Detroit to Amsterdam earlier this week I was required to board the plane not by showing a passport or boarding pass, but by pausing in front of a facial recognition camera which subsequently printed out a piece of paper with my name and seat number on it (although it appears I could have opted out, that was not disclosed by Delta Airlines staff the time). Anecdotally this gives passengers the feeling that facial recognition is robust and mature, but Adam mentions that this not the case and that removed from highly controlled environments the accuracy of recognition is closer to an abysmal 2%.

Images are only effective in these datasets when the interocular distance (the distance between the pupils of your eyes) is a minimum of 40 pixels. But over the years this minimum resolution has been moving higher and higher, with the current standard trending toward 300 pixels. The increase is not surprising as it follows a similar curve to the resolution available from digital cameras. The number of faces available in data sets has also increased along a similar curve over the years.

Adam’s talk recounted the availability of face and person recognition datasets and it was a wild ride. Of note are data sets by the names of Brainwash Cafe, Duke MTMC (multi-tracking-multi-camera),  Microsoft Celeb, Oxford Town Centre, and the Unconstrained College Students data set. Faces in these databases were harvested without consent and that has led to four of them being removed, but of course, they’re still available as what is once on the Internet may never die.

The Microsoft Celeb set is particularly egregious as it used the Bing search engine to harvest faces (oh my!) and has associated names with them. Lest you think you’re not a celeb and therefore safe, in this case celeb means anyone who has an internet presence. That’s about 10 million faces. Adam used two examples of past CCCamp talk videos that were used as a source for adding the speakers’ faces to the dataset. It’s possible that this is in violation of GDPR so we can expect to see legal action in the not too distant future.

Your face might be in a dataset, so what? In their research, Adam and Jules tracked geographic locations and other data to establish who has downloaded and is likely using these sets to train facial recognition AI. It’s no surprise that the National University of Defense Technology in China is among the downloaders. In the case of US intelligence organizations, it’s easier much easier to know they’re using some of the sets because they funded some of the research through organizations like the IARPA. These sets are being used to train up military-grade face recognition.

What are we to do about this? Unfortunately what’s done is done, but we do have options moving forward. Be careful of how you license images you upload — substantial data was harvested through loopholes in licenses on platforms like Flickr, or by agreeing to use through EULAs on platforms like Facebook. Adam’s advice is to stop populating the internet with faces, which is why I’ve covered his with the Jolly Wrencher above. Alternatively, you can limit image resolution so interocular distance is below the forty-pixel threshold. He also advocates for changes to Creative Commons that let you choose to grant or withhold use of your images in train sets like these.

Adam’s talk, MegaPixels: Face Recognition Training Datasets, will be available to view online by the time this article is published.

Take Pictures Around A Corner

August 25, 2019 by 8 Comments

One of the core lessons any physics student will come to realize is that the more you know about physics, the less intuitive it seems. Take the nature of light, for example. Is it a wave? A particle? Both? Neither? Whatever the answer to the question, scientists are at least able to exploit some of its characteristics, like its ability to bend and bounce off of obstacles. This camera, for example, is able to image a room without a direct light-of-sight as a result.

The process works by pointing a camera through an opening in the room and then strobing a laser at the exposed wall. The laser light bounces off of the wall, into the room, off of the objects on the hidden side of the room, and then back to the camera. This concept isn’t new, but the interesting thing that this group has done is lift the curtain on the image processing underpinnings. Before, the process required a research team and often the backing of the university, but this project shows off the technique using just a few lines of code.

This project’s page documents everything extensively, including all of the algorithms used for reconstructing an image of the room. And by the way, it’s not a simple 2D image, but a 3D model that the camera can capture. So there should be some good information for anyone working in the 3D modeling world as well.

Thanks to [Chris] for the tip!

Inventor And Detective Create Range Of Snack-Hiding Devices

August 25, 2019 by 8 Comments

Anyone who has had to deal with siblings, their friends, flatmates or parents who are overly fond of snacks may know this issue: you bought some snacks for your own consumption, but before you can get to them they have vanished. Naturally, nobody knows what happened to said snacks and obviously outraged that anyone would dare to do such a dastardly thing like eating someone else’s snacks.

This is the premise behind British inventor [Colin Furze]’s new series of YouTube videos (embedded after the break). Teaming up with former Scotland Yard detective [Peter Bleksley], their goal is to find ways to hide snacks around the house where curious and peckish individuals will not find them. Though a snack-company sponsored series (Walkers) and featuring snack names that will ring no bells for anyone outside of the UK, it nevertheless shows some innovative ways to hide snacks.

The first episode shows how one can hide snacks (or something else, naturally) inside a door. The second tweaks a standing lamp to add some hidden drawers, and the third episode creates a hidden compartment behind a television. Perhaps the most intriguing part of these episodes is the way it highlights how easy it is to not just hide snacks around the house, but also devices for automation and monitoring. Just think how one could use these tricks for IoT projects and the like.

Continue reading “Inventor And Detective Create Range Of Snack-Hiding Devices” →

The Danish Internet Of Hot Tubs

August 24, 2019 by 2 Comments

Every hacker camp has its own flavor, and BornHack 2019 in the Danish countryside gave us the opportunity to sample some hacker relaxation, Scandinavian style. Among the attractions was a wood-fired hot tub of gargantuan proportions, in which the tired attendee could rejuvenate themselves at 40 Celcius in the middle of the forest. A wood-fired hot tub is not the easiest of appliances to control, so to tame it [richard42graham] and a group of Danish hackerspace friends took it upon themselves to give it an internet-connected temperature sensor.

The starting point was a TMP112 temperature sensor and an ESP8266 module, which initially exposed the temperature reading via a web interface, but then collapsed under too much load. The solution was to make the raw data available via MQTT, and from that create a web interface for the event bar, Twitter and IRC bots. There was even an interface to display hot tub temperature on the ubiquitous OHMlights dotted around the camp.

It’s more normal to control a hot tub via an electric heater, but since the wood fire on this one has to be tended by a camp volunteer it made sense to use the IRC system as an alert. It will be back at BornHack 2020, so we’ll have to do our job here at Hackaday and spend a long time lounging in the hot tub in the name of journalistic research to see how well it works.

Hacked Hoverboards Become Potent RC Tank

August 24, 2019 by 11 Comments

Hoverboards were the darling, or perhaps the scourge, of the last few years, Banned by vigilant airlines, they’re a great way to break an ankle or set your house on fire. However, they’re also a treasure trove of valuable parts for hacking, as [Aaron] ably demonstrates with his RC tank build.

[Aaron’s] build utilizes not only the hoverboard’s torquey hub motors but also the original control hardware, too. This is a cinch to repurpose, thanks to the custom firmware for the original controller developed by [Lucy Fauth], whose work we have featured before.

The hacked parts are crammed into a chassis built with aluminum extrusion, and the final result is a nimble and robust tank with one motor per wheel. This enables some exciting driving dynamics. Additionally, with all the torque available, [Aaron] is even able to ride the tank like an electric skateboard.

It’s a fun build that shows off the raw power available from the hoverboard hardware. We fully expect to see these parts remain popular in the hacking scene in the coming years. Video after the break.

Continue reading “Hacked Hoverboards Become Potent RC Tank” →

Conductive Origami Lights Up Your Life

August 24, 2019 by 2 Comments

It’s taken mobile phone developers years to develop electric circuits and displays that can fold. Finally he first few have come to market — with mixed reviews and questionable utility at best. For all that R&D, there are a lot of other cases where folding circuitry might have been more useful than it seems these handsets have been. One of those is conductive origami, which in this case allows for light fixtures that turn themselves on as they are unfolded.

This conductive origami is produced by [Yael Akirav] using a 3D printer to deposit the conductive material onto fabric. From there, the light fixture can be unfolded into its final position and turned on. This isn’t just a decorative curiosity though, the design of the folding material actually incorporates the ability to turn itself on as it is unfolded. One device brightens itself as it is slowly unfolded.

This is an interesting take on foldable circuits in general, especially with some of the functionality incorporated into the physical shape of the material. We’ve seen conductive elements embroidered into fabric before, but this takes it to a new level. Surely there are more applications for a device like this that we will see in the future as well.

Thanks to [t42] for the tip!

A Colorful Way To Play Chess On An ATmega328

August 24, 2019 by 2 Comments

We’ve all seen those chess computers that consist out of a physical playing field, and a built-in computer that would indicate where you should put its pieces while inputting the position of your pieces in some way. These systems are usually found in a dusty cardboard box in a back room’s closet, as playing like this is fairly cumbersome, and a lot depends on the built-in chess computer.

This take by [andrei.erdei] on this decades-old concept involves an ATmega328p-based Arduino Pro Mini board, a nice wooden frame, and 4 WS2812-based 65×65 mm RGB 8×8 LED matrices, as well as some TTP223 touch sensors that allow one to control the on-board cursor. This is the sole form of input: using the UP and RIGHT buttons to select the piece to move, confirm with OK, then move to the new position. The chess program will then calculate its next position and indicate it on the LED matrix.

Using physical chess pieces isn’t required either: each 4×4 grid uses a special pattern that indicates the piece that occupies it.  This makes it highly portable, but perhaps not as fun as using physical pieces. It also kills the sheer joy of building up that collection of enemy pieces when you’ve hit that winning streak. You can look at the embedded gameplay video after the break and judge for yourself.

Continue reading “A Colorful Way To Play Chess On An ATmega328” →