Argos Book Of Horrors

July 26, 2020 by Bryan Cockfield 14 Comments

If you live outside the UK you may not be familiar with Argos, but it’s basically what Americans would have if Sears hadn’t become a complete disaster after the Internet became popular. While they operate many brick-and-mortar stores and are a formidable online retailer, they still have a large physical catalog that is surprisingly popular. It’s so large, in fact, that interesting (and creepy) things can be done with it using machine learning.

This project from [Chris Johnson] is called the Book of Horrors and was made by feeding all 16,000 pages of the Argos catalog into a machine learning algorithm. The computer takes all of the pages and generates a model which ties the pages together into a series of animations that blends the whole catalog into one flowing, ever-changing catalog. It borders on creepy, both in visuals and in the fact that we can’t know exactly what computers are “thinking” when they generate these kinds of images.

The more steps the model was trained on the creepier the images became, too. To see more of the project you can follow it on Twitter where new images are released from time to time. It also reminds us a little of some other machine learning projects that have been used recently to create short films with equally mesmerizing imagery. Continue reading “Argos Book Of Horrors” →

Interactive CNC Foam Cutter Churns Out Abstract Art

July 11, 2020 by Adil Malik 3 Comments

Foam is certainly an indispensable raw material for various craft and construction projects. Any serious sculptor however, inevitably grows tired of grinding through a foam block using a simple preheated utensil. The next step up, is to assemble a simple but thoroughly effective hot wire cutting contraption, formed out of a thin guitar wire held taut on a “C” shaped mounting frame. Finally, the addition of some electronics to regulate the power delivery makes this simple tool useful for most settings.

[Freddie] has taken this basic idea a step further, by building a complete multi-axis CNC foam cutter intended as an interactive exhibit on computational art. The CNC has the traditional three Cartesian axes but the platform hosting the foam piece can also rotate, introducing an additional degree of freedom. As this is indented to be controlled by attendees, there is no G-code in the mix, rather the inputs of an Xbox controller are applied directly to the work piece.

What is very interesting is how the resulting tool path is visualised and displayed. [Freddie] explains that while the user input tool path could be generated and displayed as equivalent G-code, it does not capture and convey the inherent organic nature of the finished pieces. The solution [Freddie] came up with is to display the toolpath much like a series of musical notes!

We would have loved to have a go at this machine in person, but seeing that isn’t possible in the current circumstances, you can either build a simpler machine we featured earlier or [Freddie] could perhaps fire up a camera and let us control it via the interweb, with a live video feed ofcourse!

Continue reading “Interactive CNC Foam Cutter Churns Out Abstract Art” →

Mmm… Obfuscated Shell Donuts

July 10, 2020 by Sven Gregori 6 Comments

In case you grow tired of clear-written, understandable code, obfuscation contests provide a nice change of scenery, and trying to make sense of their entries can be a fun-time activity and an interesting alternative to the usual brainteasers. If we ever happen to see a Simpsons episode on the subject, [Andy Sloane] has the obvious candidate for a [Hackerman Homer] entry: a rotating ASCII art donut, formatted as donut-shaped C code.

The code itself actually dates back to 2006, but has recently resurfaced on Reddit after [Lex Fridman] posted a video about it on YouTube, so we figured we take that chance to give some further attention to this nifty piece of art. [Andy]’s blog article goes in all the details of the rotation math, and how he simply uses ASCII characters with different pixel amounts to emulate the illumination. For those who prefer C over mathematical notation, we added a reformatted version after the break.

Sure, the code’s donut shape is mainly owed to the added filler comments, but let’s face it, the donut shape is just a neat little addition, and the code wouldn’t be any less impressive squeezed all in one line — or multiple lines of appropriate lengths. However, for the actual 2006 IOCCC, [Andy] took it a serious step further with his entry, and you should definitely give that one a try. For some more obfuscated shell animations, check out the fluid dynamics simulator from a few years back, and for a more recent entry, have a look at the printf Tic Tac Toe we covered last month.

Continue reading “Mmm… Obfuscated Shell Donuts” →

Live Map Of London Tube Created In PCB And Lights

July 5, 2020 by Bryan Cockfield 20 Comments

If you’re a frequent traveler on a public transit system, it can be helpful to know when the trains or buses are arriving and if there are any delays. We might reach for a tablet to mount on the wall, but that relies on keeping the OS, the software, and its library dependancies up to date. For true reliability you’ll need to build directly in hardware, which is exactly what this map of the London tube system uses.

The base map is printed directly on PCB, with LEDs along each of the major routes to indicate the current location of the trains. A few small chips handle the WiFi connection — it appears to our eye to be an ESP8266 — and pulling the information about the trains from the London Underground API (it would be virtually impossible to build everything for this project in hardware). The hardware can be easily reprogrammed, and with the PCB layout this could be adapted for other public transit fairly easily.

Even apart from the philosophical differences on design between hardware and software approaches, we still appreciate the aesthetic of LEDs on PCB. In fact, we’ve seen a whole host of artwork on PCBs ever since the price came down dramatically in the past two decades.

Thanks to [Al] for the tip!

LED Art Hack Chat

June 29, 2020 by Dan Maloney 1 Comment

Join us on Wednesday, July 1 at noon Pacific for the LED Art Hack Chat with Aaron Oppenheimer!

From the first time humans crawled into a cave with a bit of charcoal to sketch scenes from the world around them, artists have been searching for new media and new ways to express themselves. Natural products ruled for thousands of years, with pigments stolen or crafted from nature as well as wood, ivory, bone, and stone for carving. Time and experience guided our ancestors to new and better formulations and different materials, to the point that what qualifies as art and what we’d normally think of as technology have, in many cases, blended into one, with the artist often engineering projects of mammoth proportions and breathtaking beauty.

Aaron Oppenheimer co-founded color+light, a company that specializes in large-scale custom art installations for companies like Google, Nike, and Nissan. One of their projects, the “Oddwood Tree”, is displayed alongside other gigantic art pieces at Area15 on the Las Vegas strip. His most recent project, fluora, is a digital houseplant, with addressable LEDs in the leaves that can be controlled by a smartphone app or respond to stimuli in the environment.

Aaron will join us on the Hack Chat to discuss the LED as artistic medium. Join us as we learn what it takes to make enormous art that’s strong enough to interact with yet responsive enough to be engaging.

Our Hack Chats are live community events in the Hackaday.io Hack Chat group messaging. This week we’ll be sitting down on Wednesday, July 1 at 12:00 PM Pacific time. If time zones have you down, we have a handy time zone converter.

Click that speech bubble to the right, and you’ll be taken directly to the Hack Chat group on Hackaday.io. You don’t have to wait until Wednesday; join whenever you want and you can see what the community is talking about.

Continue reading “LED Art Hack Chat” →

Recreating Paintings By Teaching An AI To Paint

June 27, 2020 by Maya Posch 9 Comments

The Timecraft project by [Amy Zhao] and team members uses machine learning to figure out a way how an existing painting may have been originally been painted, stroke by stroke. In their paper titled ‘Painting Many Pasts: Synthesizing Time Lapse Videos of Paintings’, they describe how they trained a ML algorithm using existing time lapse videos of new paintings being created, allowing it to probabilistically generate the steps needed to recreate an already finished painting.

The probabilistic model is implemented using a convolutional neural network (CNN), with as output a time lapse video, spanning many minutes. In the paper they reference how they were inspired by artistic style transfer, where neural networks are used to generate works of art in a specific artist’s style, or to create mix-ups of different artists.

A lot of the complexity comes from the large variety of techniques and materials that are used in the creation of a painting, such as the exact brush used, the type of paint. Some existing approaches have focused on the the fine details here, including physics-based simulation of the paints and brush strokes. These come with significant caveats that Timecraft tried to avoid by going for a more high-level approach.

The time lapse videos that were generated during the experiment were evaluated through a survey performed via Amazon Mechanical Turk, with the 158 people who participated asked to compare the realism of the Timecraft videos versus that of the real time lapse videos. The results were that participants preferred the real videos, but would confuse the Timecraft videos for the real time lapse videos half the time.

Although perhaps not perfect yet, it does show how ML can be used to deduce how a work of art was constructed, and figure out the individual steps with some degree of accuracy.

Continue reading “Recreating Paintings By Teaching An AI To Paint” →

Flexible PCB Earrings Put The Art In Art Deco

May 24, 2020 by Kerry Scharfglass 5 Comments

Earrings have been a hackers’ target for electronic attachment for quite a while, but combining the needed components into a package small enough to wear in that finicky location is quite a challenge. If [Sawaiz Syed]’s Art Deco Earrings are anything to go by, ear computers have a bright future ahead of them!

This is a project unusually well described by its name. It is in fact an earring, with art deco styling. But that sells it way too short. This sliver of a flex circuit board is double sided to host an ATtiny, accelerometer, LDO, and eight 2020 formfactor controller-integrated LEDs. Of course it’s motion sensitive, reacting to the wearer’s movement via LED pattern. [Sawaiz] makes reference to wearing it while dancing, and we can’t help but imagine an entire ballroom all aglow with tiny points of LED light.

The Art Deco Earrings are also set apart by the thoroughness of their documentation (have we mentioned how much we love detailed documentation?). [Sawaiz] not only drops the source in your lap, but the README in the Github repo linked at the top walks the reader through each component of the design in detail. Plus the PCBA render is so complete it includes a model of the wire loop to fit through the wearer’s ear; how cool is that? The single piece that’s still in progress is the battery. The earring itself hosts an LDO, so all that is required is stashing a battery somewhere discrete, perhaps in the user’s hair? We’re looking forward to seeing what [Sawaiz] works out.

For the full effect, check out the gif of an assembled unit in action after the break.

Continue reading “Flexible PCB Earrings Put The Art In Art Deco” →