Hello everyone, and welcome back to another weekly design challenge summary. It’s the sixth week of the PiCasso Design Challenge. Lots of progress has been made over the past week, but before we get to the good stuff, let’s take a few moments to learn more about the challenges, their challengers, and what hardware they are using.
PiCasso Design Challenge
About The Challenge
Featured as the first Design Challenge of 2019, this challenge task its twenty participants with creating their artistic masterpiece using a Raspberry Pi Model 3B+. Challengers can either push their boundaries by using the computing power of Pi as the brains of their artistic work or to improve the aesthetics of their surroundings.
While twenty project proposals were chosen to be the official projects of the challenge, entering the challenge is not limited to just those twenty community members. Anyone can join the Challenge as a non-sponsored Challenger, and still be eligible win one of the three prizes. If you are a non-sponsored challenger all you need to do is integrate the components from the official challenger kit into your design, and post 10-weekly blogs detailing your project’s build.
The Official Kit
The PiCasso Design Challenge is sponsored by the Raspberry Pi Foundation. Each challenger will receive all of the components below plus a wide assortment of additional passive, and digital components to build their projects with. If you would like to enter the challenge for yourself, or if you just want to follow along at home, you can purchase all of the components in the kit right here at Element14.
To learn more about each of these components as well as the other components included in the kit, visit the official kit announcement page or click the links above to purchase them for your project.
- Grand Prize: DJI Mavic 2 Pro Videography Drone
- Runner Up Prize: Nikon D3500 DSLR & Lens Kit
- Finisher Prize: Raspberry Pi Arcade Kit.
To be eligible, a challenger has to complete their project, use products from the challenger kit, and post at least 10 updates in the Sixth Sense Design Challenge space.
To learn more about the prizes, head over to the PiCasso Design Challenge Prize Page.
The Past Week In Review
Over the past several days, April 14th - April 20th, we have had a total of six updates posted across six projects. This week, I will be highlighting three of my favorite updates from this past week, but before I get to that, let's take a look at which projects were updated.
- Project:Power Pi Image Processing - by Brenda Armour (armour999)
- Project:Art-a-tronic exhibition: Mannequin - by Enrico Miglino (balearicdynamics)
- Project:Adapting Art to Ages - by Frank Milburn (fmilburn)
- Project:Colorful Rotating Chinese Lantern - by Vincent Wong (wesee)
- Project:Artificial Intelligence powered CNC Painting Machine - by Attila Tőkés (bluetiger9)
- Project:Tableau - by Eric Jorgensen (jorgy)
This Week’s Top Updates
Project: Art-a-Tronic - PiCasso Design Challenge Blog #10 - The MDK in Practice and The Eye Light
Enrico Miglino claims the top spot again this week with another great update to project Art-a-Tronic. In the project’s tenth update he puts the Mannequin Development Kit into action by duplicating the hardware setup at home in his laboratory, but that is not the main focus of the update. The post also focuses on solving the incompatibility issues with the NeoPixel library when using the Stepper Library in the same software stack. “Despite the incompatibility problem of the NeoPixel library with the Stepper library as discussed in Episode 6, Seeing the solution I have adopted now I am almost sure that in this particular context the NeoPixel ring was not the right choice,” he wrote. “The first approach was to use three other pins of the Arduino and a shift register. I have also developed in past a Multiple Shift-Out Registers for Arduino, including the board and PCB. Some of them are already here around and I have plenty (some hundred) of shift registers, so why not? The answer is simple: because as much a solution is simple, as much it is elegant. I am not a fanatic of the hardware redundancy and if I can reach the same result with less stuff, the solution is better.”
The update covers much more than I have room to write about here including software updates, new hardware additions, and much more, so head over to the full post by clicking the link above to check it out.
Project: Adapting Art to Viewers - PiCasso Design Challenge Blog #7 - New Art
One thing I am very passionate about is involving children in my projects. My nephew is well on his way to becoming a competent maker because I involve him in so many of my projects, and that is why the seventh update to Frank Milburn’s project, Adapting Art to Viewers, caught my eye this week. Utilizing the free labor force readily available at his disposal (his grandchildren) he had them create a few drawings he would be able to use as data sets when training the AI portion of his project. “This week I have involved my grandchildren in making art on the Pi, or at least Pi related art. These three short visuals with musical accompaniment will be played when the child who created the art is recognized. We also fooled around with the image recognition which while improving is unlikely to ever get high accuracy, especially at higher frame rates,” he wrote.
While his 5-year old granddaughter seems to have a predisposition to becoming a naturist, his 3-year old granddaughter might very well be the next Einstein or Hawking, but one thing is for certain, his 7-year old grandson will be the creator of our future robot overlords. In all seriousness though, the work Frank is doing on this project is pretty interesting, and I am quite excited to see how the project comes together in the coming weeks! Check out the full post for yourself at the link above!
Project: AI powered CNC Painting Machine - PiCasso Design Challenge Blog #2 - Image Stylization / Style Transfer
Up last this week (but definitely not least) is project AI Powered CNC Painting Machine by Attila Tőkés. In the project’s second update, Attila explores methods to “extract” image stylization from existing images, and then use an algorithm to recreate existing, unrelated images in the style of the original image that was originally scanned. “Image stylization/style transfer is a process the can be described as "extracting the style" from an image and "applying it" to a second image,” Attila wrote. “An example would be to take a painting (ex take Vincent van Gogh's The Starry Night) and apply to a photo (Stanford campus).” Two methods were explored including: “A Neural Algorithm of Artistic Style,” and “Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis.” Both of which are well above my current understanding of AI. If you would like to know more about these processes, or a few more “Raspberry Pi” friendly versions of similar algorithms, head over to the full post at the link above.
That is going to wrap up this weeks coverage of the Picasso Design Challenge. If you have not yet done so, please take a moment and check out all of this week's updates, and show our challengers some love in the comments section of each update post. Check back next week for another weekly design challenge summary.